Engineering, in contrast, is the technique of using science to produce tools we can consistently use in the world.
Semantically, wouldn't one call a neanderthaler creating/perfecting a useful tool an engineer as well, even though science was not really used directly?
Chomsky said some interesting things about science’s impact on engineering/medicine:
“Until very recently engineers learned from the sciences, but most of what they knew was craft. The crafts were so much more advanced than theoretical understanding that the engineers worked like artists. You know, you learned how to do it. Physics didn’t really get to contribute to engineering theory until fairly recently. In fact, when I got to MIT not that long ago in the 1950s, it was still largely an engineering school, and physics was taught as a service. But if engineers wanted to construct electric circuits or build a bridge and so on, you learned for the profession and you learned some physics – and it helped you, but now it’s changed.” (http://www.justresponse.net/chomsky_offers_advice.html)
Science is a social movement, with a history we can look at. Like many movements, there’s lots misleading written about it. Often going into religion, especially when groups try to claim its prestige for themselves. (We might include mainstream economics here, but hilariously many capitalist economists aim beyond science, closer to math. At least many Marxists "only" claimed to have Scientific Truths.)
>At least many Marxists "only" claimed to have Scientific Truths.)
Marx was probably using the German word "Wissenschaft"[1] which is not the same as modern empirical science, it's means something like structured knowledge and it includes branches of knowledge like history or literature that aren't called science in English.
For all we know, our current version of science could be as off the mark to some possible formulation as neanderthal's version of the world is to us.
Engineering works on top of what 'model' is available. Engineering even works when the model is incomplete but made irrelevant (not knowing the correct model of quantum mechanics didn't slow down the industrial revolution.)
Science is a great tool to improve greatly our understanding about how the world works, but it's not necessary to know SOME facts.
Stone age people were smart, and they knew that things go down, some materials are harder than others and a pointing object hurts.
Engineering is the process of using that to build a construction like Newgrange or a spear to hunt. Making use of your understanding of the world to take advantage of the medium.
Science just allows you to greatly increase the understanding of the world, in a more precise manner and at a faster pace. You learn properties of rocks (science) that can be used to infer newer materials, lighter and more resistant than others (engineering), or to predict with more precision star movements (science) and allow you better navigation through sea (engineering).
I think people get hung up on the idea of applied science, since applying science is typically one of the best methods of engineering. Our ability to engineer things is dramatically improved by applying the knowledge, techniques, and processes provided by science. But it's not the only way.
I feel the article dances around what the fundamental purpose of engineering is - that is to create viable real-world objects by solving problems of constraints and tradeoffs. Nearly all meaningful engineering problems can be posed as a series of constraints and tradeoffs. Even if you were out trying to sharpen a stick into a spear, you're intuitively (and probably iteratively) solving the problem of constraints and tradeoffs.
When you set out the make the spear, you might not even know the full extent of the problem. You know that at a high level the spear (assume a throwing spear) must be long enough to be stable in flight, short enough to not be a burden, heavy enough to fly well and deliver energy on impact, light enough to throw, thin though to grip easily, but thick enough to not shatter on impact, pointy enough to inflict damage, but not too pointy for the point to be too fragile.
You now get to solve this problem. Sometimes as an engineer you might get the exact values of these constraints given to you, sometimes you have to go find them, and sometimes you wing it. You can use science (work from material properties, dynamics, deformation, biomechanics), you can use some 'direct empirical evidence' (screw around and guess and check =D ) and iterate through a bunch of spear designs, or some rules of thumbs that your uncle told you or a mix of all.
The most important thing is that you are managing trade-offs and constraints at some level.
Actually, this is where I stopped reading because in my opinion the author gets it exactly backwards.
Engineers need to build things with safety thresholds. Most devices will probably experience situations where it is accelerating at a greater rate than just that due to gravity. Engineers work in a messy world without controls and try their best to deal with non-ideal scenarios. They live in a world where a cow often cannot be idealized as a sphere. Because of this, the precise acceleration often doesn't matter, what matters is that the device works exactly within some range of operating environments. Only in cases of metrology do engineering devices need precise measures of external factors.
Meanwhile, scientists need precise measurements of environmental influences because they seek to eliminate them from experiments. otherwise, the signal cannot be distinguished from the noise. That is why the mass of an electron is known to an exaggerated precision, not because an engineer needs to build a device for it, but because (gross simplification) the precision is needed to be able to subtract electron related signals from the Higgs boson signal.
Well, IC manufacturing may be the GP's perfect example.
You can't count on wafers being exactly at the same size, so you build your machinery in a way that precise sizes aren't important, only relative measures within a wafer is.
You can't count on alignment measurements, so you build your machines to use the only alignment invariant available, that is the size of a single wafer (and call it self-aligning).
You can't count on features having any specific size. They vary wildly, in proportions that in any other specialization of engineering would be disastrous. So, you just manufacture them, test after the fact, and throw away the parts that do not comply.
The entire field of IC manufacturing is about how to not need measurements and be resilient to errors.
Also, I don't know much about nuclear reactor design, but I'd be very surprised if they used measurements with several algarisms.
All engineering projects start with unknowns - the specifics of the requirements, and also external constraints such as the geology of proposed dam or bridge sites - but answering them rarely makes any significant addition to the body of scientific knowledge.
In this regard, the practice of software development (as opposed to the state of the art) is mostly distinguished by an aversion to seeking answers to these questions before developing a solution. Software engineering attempts to change this, but in practice, it is often pursued through ad-hoc rules-of-thumb of the sort that craftsmen use, rather than anything equivalent in rigor to physical engineering's foundation in the physical sciences.
The easiest way I found to describe the differences between science and engineering (and I think about this fairly often) is by analogy with learning how to drive a (manual) car.
There are a few key skills one needs to master to be able to drive a car through any kind of circuit: from the basics, such as shifting up and down and the appropriate times, to drifting at will. Given enough time and "research", people can master this aspect effectively. That's science.
Engineering is the rest of the driving experience. The placement of traffic lights, how to synchronise them so two perpendicular roads are not green at the same time, and all of the signaling (stop, yield, etc) that we are used to.
The difference is there is a bigger solution space in engineering: there is no "one correct answer", and the factors involved in determining whether something is a good engineering solution are much broader. For example, reusing the signaling example above, the "stop" sign doesn't have to be a hexagonal red shape, but some early civil engineer figured that shape is the most adequate.
I don't really know, computer science seems concerned with equipping the student with the best tools to solve a problem, and mastery in the creation of them. Computer engineering concerns itself with similar concepts, but at a lower level of abstraction. It's still not EE, which is lower still.
I studied computer/electrical engineering for part of my undergrad, and computer science for my masters. The biggest difference I recognized was the level of abstraction. Each has it's own maths, systems of reasoning and analysis.
"Computer science" courses are mostly misnamed. The research faculties they're located in are usually scientific (in that they concern research into computability and computation), however what undergraduate students mostly learn is the primitive form of engineering that we've gleaned from that science so far.
I have an MEng in Computing, because my alma mater is one of the few that realises the distinction.
I'm talking from a graduate perspective, but the courses I TAd for undergrad were not what I would call 'primitive engineering' as there is an emphasis on analysis and rigor.
It is a science to be able to build programs and prove things about them. It is way more abstract maths than anything in engineering (linear algebra still has some physical grounding in reality, where category theory is more for the organization of mind).
The quote consists of seven words, where a word in this context is a lexical unit that consists of alphanumeric characters, and the words are separated by some kind of whitespace.
Generally accepted philosophy of science is positivism, which states that 'why' in general is unanswerable and we should satisfy ourselves with descriptions. Therefore science is more about precise descriptions and their predictive power.
Science is commonly defined as usage of scientific method. So science is just trying to find truth with the best possible tools. Nothing more and nothing less. I think here is the common and problematic misconception. Scientists are saying "look, we used the best tools" and people are hearing "hey, we found the truth!". But science is not about the destination, it's all about the journey.
Engineering lacks such rigid definition. I've thought about it a lot, and I think deep down engineering is always optimization.
If you invent a new need for people and fill it with a toy, that would be marketing. But if you solve previously solved problem better, that's engineering. You got closer to the optimum, even though the way you solved might be described as "invention". So engineering inventions would be just leaps in optimization.
Now if you optimize economical or cultural stuff, that would not be engineering. So engineering is "optimization of technology".
I do not think "engineering" as a term should be expanded to the fields this article suggests. The term started off as "art of building of siege engines". And so far we don't have "sales-engineering". It's called marketing and it works pretty good considering there is big human factor everywhere. Medicine is good term too. These fields are just naturally more difficult to squeeze accurate data out.
I think the best model how to deal with the human factor comes from medicine. Rigorous meta-analysis is pretty solid. If we import the way tech or natural science is researched, that would be: "just use better instruments and measure more carefully". Which is pretty obvious dead. Because you will just get more expensive studies that still disagree with each other.
I don't think it's quite that clear cut. Crash tests are focused on building safer cars, but it also neededs to validate it's model of what happens in a crash. So, sending human cadavers down a track is science, but what about comparing the crumple zone deformation with computer models?
Surely engineering is just about building stuff that works, or "meets requirements".
Is there any kind of engineering for which that is basically not the case?
Science is about understanding the natural world by building and testing hypothesis.
I can think of some grey areas. For example, in the 20th century it was discovered that the harmonic modes of oscillation of large structures, like bridges, was rather important to their structural stability. Once this was well understood, clearly it was the engineers who built better and stronger structures with inbuilt dampening and so forth. But who actually sat down and figured out the math behind structural harmonic response, and showed how it could lead to disaster? A scientist or an engineer? It seems both very sciency and engineery.
There's a grey area, but I find your example (or the way you posed it) a bad one. Creating a mental model is science, even if it was done to be applied in a engineering context.
First, even the math and science behind the structural harmonic response was made by an engineer, it was a science advance after all. They allowed people to better understand the world, creating models of phenomena that are observable. Of course, engineers can do science, as anybody else can, and that does not transform science endeavours into engineering endeavours.
However, it must be noted that there is a lot of science in engineering: people can approach engineering problems and subjects with a science mindset so they can advance engineering. A new DSP algorithm is basically math, even if it is math with some constraints set by engineering realities, and may be analysed as such - more often than not, by people with engineering degrees, who are doing science.
Both engineering and science are less about fields of knowledge and more about process: if you are trying to understand something, and to observe things that exist in a methodical way, aiming to create a model that can reproduce and predict events, you are doing science. If you are trying to design something that help you efficiently achieve some practical goal, you are doing engineering.
I think the areas of signal processing, machine learning, information theory, communications, numerical techniques, etc. straddle the science-engineering border on both sides.
While the humorous distinctions are all great, they are also sort of tiresome to me. Maybe I'm the only one. To me it's a symptom of a world where we have too much specialization. To do "science" a scientist will often have to do "engineering" and I also feel like engineering expands knowledge just like science does.
I think this is missing the point about resource management, which I think it's one of the most important things about engineering. That is, the difference between "effective" and "efficient"
Author makes a really good point but I think he presents it in a way that doesn't make it obvious and sets it up as ego stroking "science vs engineering" (which I can already see in the comments).
Science on it's own is not robust enough to build models for real-world decision making - things we could consider in the domain of engineering. If a scientific paper finds a correlation between red wine and heart disease this may be interesting and important clue for scientific study but this is not enough to develop a model of nutrition and certainly not enough to recommend consuming red wine for hearth health. Plenty of people don't realize this and you see articles "red wine good for your heart" - without a robust model of human metabolism/nutrition we just can't make claims like this. The food pyramid is a perfect example of this flaw.
Definitely true about science on it's own not being robust enough -- but I think you might be going a bit too far saying that "without a robust model of human metabolism/nutrition we just can't make claims like this". Yes, such a model would be awesome, but since it's not forthcoming any time soon, it is still useful in the meantime to be able to make claims of this nature using other evidence, such as experiments and even statistical studies, as long as we don't take results for 100% truth (although, taking model results for 100% truth would not be a good idea either).
Where I work we have everything from pure science to pure engineering and it's definitely a continuum. Sometimes it's painful to be caught in the middle but I think that's really where the interesting engineering happens.
However, there are also some forms of engineering that are not very science based, you get a process or product to work and you don't exactly know why. You can still use scientific methods to get to a safer, more controlled process but don't necessary exactly know why something works. We call that "taking the art out of engineering" and it's an important part, especially in large scale manufacturing.
An interesting topic. I agree the two terms are usually imprecisely joined. But I think the author's definition is too binary.
Science and engineering are a continuum. At one end is the theoretical scientist who discovers new knowledge; at the other the engineer who applies knowledge to serve some practical end. But between them lies the applied scientist whose purpose is to extend scientific knowledge to make it valuable to scientists working in other domains. And the process of doing that always involves lots of engineering -- discovering the reproducible limits of a phenomenon or a technique toward achieving some goal. Inevitably then, many applied scientists and engineers share the same job.
For example: Were Salk and Sabin scientists or were they engineers when they refined a method to inoculate against polio? Certainly they were applying science, but what was their principal role: a) discovering new knowledge or b) refining a technique? Once it moves from the petri dish to the human, isn't medicine usually more like engineering than science?
As I see it, at some point the boundary between science vs engineering blurs to an a degree that the difference is negligible.
56 comments
[ 2.9 ms ] story [ 109 ms ] threadSemantically, wouldn't one call a neanderthaler creating/perfecting a useful tool an engineer as well, even though science was not really used directly?
“Until very recently engineers learned from the sciences, but most of what they knew was craft. The crafts were so much more advanced than theoretical understanding that the engineers worked like artists. You know, you learned how to do it. Physics didn’t really get to contribute to engineering theory until fairly recently. In fact, when I got to MIT not that long ago in the 1950s, it was still largely an engineering school, and physics was taught as a service. But if engineers wanted to construct electric circuits or build a bridge and so on, you learned for the profession and you learned some physics – and it helped you, but now it’s changed.” (http://www.justresponse.net/chomsky_offers_advice.html)
Science is a social movement, with a history we can look at. Like many movements, there’s lots misleading written about it. Often going into religion, especially when groups try to claim its prestige for themselves. (We might include mainstream economics here, but hilariously many capitalist economists aim beyond science, closer to math. At least many Marxists "only" claimed to have Scientific Truths.)
Marx was probably using the German word "Wissenschaft"[1] which is not the same as modern empirical science, it's means something like structured knowledge and it includes branches of knowledge like history or literature that aren't called science in English.
[1] https://en.wikipedia.org/wiki/Wissenschaft
Stone age people were smart, and they knew that things go down, some materials are harder than others and a pointing object hurts. Engineering is the process of using that to build a construction like Newgrange or a spear to hunt. Making use of your understanding of the world to take advantage of the medium.
Science just allows you to greatly increase the understanding of the world, in a more precise manner and at a faster pace. You learn properties of rocks (science) that can be used to infer newer materials, lighter and more resistant than others (engineering), or to predict with more precision star movements (science) and allow you better navigation through sea (engineering).
I think people get hung up on the idea of applied science, since applying science is typically one of the best methods of engineering. Our ability to engineer things is dramatically improved by applying the knowledge, techniques, and processes provided by science. But it's not the only way.
I feel the article dances around what the fundamental purpose of engineering is - that is to create viable real-world objects by solving problems of constraints and tradeoffs. Nearly all meaningful engineering problems can be posed as a series of constraints and tradeoffs. Even if you were out trying to sharpen a stick into a spear, you're intuitively (and probably iteratively) solving the problem of constraints and tradeoffs.
When you set out the make the spear, you might not even know the full extent of the problem. You know that at a high level the spear (assume a throwing spear) must be long enough to be stable in flight, short enough to not be a burden, heavy enough to fly well and deliver energy on impact, light enough to throw, thin though to grip easily, but thick enough to not shatter on impact, pointy enough to inflict damage, but not too pointy for the point to be too fragile.
You now get to solve this problem. Sometimes as an engineer you might get the exact values of these constraints given to you, sometimes you have to go find them, and sometimes you wing it. You can use science (work from material properties, dynamics, deformation, biomechanics), you can use some 'direct empirical evidence' (screw around and guess and check =D ) and iterate through a bunch of spear designs, or some rules of thumbs that your uncle told you or a mix of all.
The most important thing is that you are managing trade-offs and constraints at some level.
Software engineering must be a special case. Half of it is about handling uncertain states and preparing for unknown future uses.
Engineers need to build things with safety thresholds. Most devices will probably experience situations where it is accelerating at a greater rate than just that due to gravity. Engineers work in a messy world without controls and try their best to deal with non-ideal scenarios. They live in a world where a cow often cannot be idealized as a sphere. Because of this, the precise acceleration often doesn't matter, what matters is that the device works exactly within some range of operating environments. Only in cases of metrology do engineering devices need precise measures of external factors.
Meanwhile, scientists need precise measurements of environmental influences because they seek to eliminate them from experiments. otherwise, the signal cannot be distinguished from the noise. That is why the mass of an electron is known to an exaggerated precision, not because an engineer needs to build a device for it, but because (gross simplification) the precision is needed to be able to subtract electron related signals from the Higgs boson signal.
Have s look at an IC engine and the tolerances required for an engine to work.
Let alone say correctly designing nuclear rectors to handle two phase flow in its cooling loops.
You can't count on wafers being exactly at the same size, so you build your machinery in a way that precise sizes aren't important, only relative measures within a wafer is.
You can't count on alignment measurements, so you build your machines to use the only alignment invariant available, that is the size of a single wafer (and call it self-aligning).
You can't count on features having any specific size. They vary wildly, in proportions that in any other specialization of engineering would be disastrous. So, you just manufacture them, test after the fact, and throw away the parts that do not comply.
The entire field of IC manufacturing is about how to not need measurements and be resilient to errors.
Also, I don't know much about nuclear reactor design, but I'd be very surprised if they used measurements with several algarisms.
In this regard, the practice of software development (as opposed to the state of the art) is mostly distinguished by an aversion to seeking answers to these questions before developing a solution. Software engineering attempts to change this, but in practice, it is often pursued through ad-hoc rules-of-thumb of the sort that craftsmen use, rather than anything equivalent in rigor to physical engineering's foundation in the physical sciences.
There are a few key skills one needs to master to be able to drive a car through any kind of circuit: from the basics, such as shifting up and down and the appropriate times, to drifting at will. Given enough time and "research", people can master this aspect effectively. That's science.
Engineering is the rest of the driving experience. The placement of traffic lights, how to synchronise them so two perpendicular roads are not green at the same time, and all of the signaling (stop, yield, etc) that we are used to.
The difference is there is a bigger solution space in engineering: there is no "one correct answer", and the factors involved in determining whether something is a good engineering solution are much broader. For example, reusing the signaling example above, the "stop" sign doesn't have to be a hexagonal red shape, but some early civil engineer figured that shape is the most adequate.
I studied computer/electrical engineering for part of my undergrad, and computer science for my masters. The biggest difference I recognized was the level of abstraction. Each has it's own maths, systems of reasoning and analysis.
I have an MEng in Computing, because my alma mater is one of the few that realises the distinction.
It is a science to be able to build programs and prove things about them. It is way more abstract maths than anything in engineering (linear algebra still has some physical grounding in reality, where category theory is more for the organization of mind).
"Trendy teaching as confusing science for engineering"
(I'm not a native English speaker)
The quote consists of seven words, where a word in this context is a lexical unit that consists of alphanumeric characters, and the words are separated by some kind of whitespace.
Better?
In short, hypotheses are the key instruments to investigate into unknown(and hopefully, consistent) world.
:)
More verbosely: Science is about creating knowledge about out how things work, engineering is applying knowledge to make things work.
Engineering is how you build a bridge. Science is how you find out how to tell whether that bridge will hold.
Engineering lacks such rigid definition. I've thought about it a lot, and I think deep down engineering is always optimization.
If you invent a new need for people and fill it with a toy, that would be marketing. But if you solve previously solved problem better, that's engineering. You got closer to the optimum, even though the way you solved might be described as "invention". So engineering inventions would be just leaps in optimization.
Now if you optimize economical or cultural stuff, that would not be engineering. So engineering is "optimization of technology".
I do not think "engineering" as a term should be expanded to the fields this article suggests. The term started off as "art of building of siege engines". And so far we don't have "sales-engineering". It's called marketing and it works pretty good considering there is big human factor everywhere. Medicine is good term too. These fields are just naturally more difficult to squeeze accurate data out.
I think the best model how to deal with the human factor comes from medicine. Rigorous meta-analysis is pretty solid. If we import the way tech or natural science is researched, that would be: "just use better instruments and measure more carefully". Which is pretty obvious dead. Because you will just get more expensive studies that still disagree with each other.
Surely marketing is creating markets. You have a thing, and you make people want to buy it, thus creating a market. Marketing.
Is there any kind of engineering for which that is basically not the case?
Science is about understanding the natural world by building and testing hypothesis.
I can think of some grey areas. For example, in the 20th century it was discovered that the harmonic modes of oscillation of large structures, like bridges, was rather important to their structural stability. Once this was well understood, clearly it was the engineers who built better and stronger structures with inbuilt dampening and so forth. But who actually sat down and figured out the math behind structural harmonic response, and showed how it could lead to disaster? A scientist or an engineer? It seems both very sciency and engineery.
First, even the math and science behind the structural harmonic response was made by an engineer, it was a science advance after all. They allowed people to better understand the world, creating models of phenomena that are observable. Of course, engineers can do science, as anybody else can, and that does not transform science endeavours into engineering endeavours.
However, it must be noted that there is a lot of science in engineering: people can approach engineering problems and subjects with a science mindset so they can advance engineering. A new DSP algorithm is basically math, even if it is math with some constraints set by engineering realities, and may be analysed as such - more often than not, by people with engineering degrees, who are doing science.
Both engineering and science are less about fields of knowledge and more about process: if you are trying to understand something, and to observe things that exist in a methodical way, aiming to create a model that can reproduce and predict events, you are doing science. If you are trying to design something that help you efficiently achieve some practical goal, you are doing engineering.
- Science finds the new things
- Research plays around with these things and susses out repeatable factors
- Development sees if those repeatable factors can be turned into a product
- Engineering builds those products
(- and sales and marketing sells them)
Engineering = applied Science
[1] A good starting point https://en.wikipedia.org/wiki/Hoare_logic
Science on it's own is not robust enough to build models for real-world decision making - things we could consider in the domain of engineering. If a scientific paper finds a correlation between red wine and heart disease this may be interesting and important clue for scientific study but this is not enough to develop a model of nutrition and certainly not enough to recommend consuming red wine for hearth health. Plenty of people don't realize this and you see articles "red wine good for your heart" - without a robust model of human metabolism/nutrition we just can't make claims like this. The food pyramid is a perfect example of this flaw.
However, there are also some forms of engineering that are not very science based, you get a process or product to work and you don't exactly know why. You can still use scientific methods to get to a safer, more controlled process but don't necessary exactly know why something works. We call that "taking the art out of engineering" and it's an important part, especially in large scale manufacturing.
Science and engineering are a continuum. At one end is the theoretical scientist who discovers new knowledge; at the other the engineer who applies knowledge to serve some practical end. But between them lies the applied scientist whose purpose is to extend scientific knowledge to make it valuable to scientists working in other domains. And the process of doing that always involves lots of engineering -- discovering the reproducible limits of a phenomenon or a technique toward achieving some goal. Inevitably then, many applied scientists and engineers share the same job.
For example: Were Salk and Sabin scientists or were they engineers when they refined a method to inoculate against polio? Certainly they were applying science, but what was their principal role: a) discovering new knowledge or b) refining a technique? Once it moves from the petri dish to the human, isn't medicine usually more like engineering than science?
As I see it, at some point the boundary between science vs engineering blurs to an a degree that the difference is negligible.