To be clear, by knowledge I do not mean regurgitating textbook definitions. What I want demonstrated is a practical usage of some bit of knowledge. [Borrowing from another post] Rather than ask:
Q. What is the Liskov substitution principle?
I would probably ask something like:
Q. A Square class is derived from a Rectangle class. Function foo has a non-const parameter of type Rectangle. Function foo is called using a variable of type Square. Is this safe, if not why?
You ask me what a Liskov substitution is in an interview as a serious primary question and not just as trivia, I'm going to give you a deer in headlights. Afterwards, I'm going to look it up, find out what it was, and even if your company gave me a job offer I'm going to reject it because there's no way in hell I want to work with people who would ask obscure trivia as serious job interview questions. I would never fault a person for not knowing what Levenshtein distance was off the top of their heads.
After a lead engineer gave me a written programming test with a lot of such trivia questions I was back in the project managers office. He asked me how things went. I said OK, I skipped some questions since they were trivia easily looked up in references, things I have not needed to memorize since taking an exam in college. I then went on to describe why their test was flawed and how a better test would go about things. His response, "you're right, the test does sound like something just slapped together in a half assed fashion." I received an offer a few days later.
A perfectly fine answer is "No, I've never heard of it before. What is it?" Then we can have a conversation.
Perfect. Sadly, then you write the following...
I'm not looking for trivia experts... A lot of my questions are to probe whether they have bothered to read books that are not typically part of the college curriculum but are important references for a professional software engineer.
Perhaps a beginning software engineer. As one with 30+ years of experience I might look at a summary of Martin's SOLID topic and think "not much new here", low priority or pass.
"S Single responsibility principle
a class should have only a single responsibility (i.e. only one potential change in the software's specification should be able to affect the specification of the class)
O Open/closed principle
“software entities should be open for extension, but closed for modification.”
L Liskov substitution principle
“objects in a program should be replaceable with instances of their subtypes without altering the correctness of that program.” See also design by contract.
I Interface segregation principle
“many client-specific interfaces are better than one general-purpose interface.”
D Dependency inversion principle
one should “depend upon abstractions, [not] concretions.”" https://en.wikipedia.org/wiki/...
I find it interesting that so many latched on the Liskov question -- and read their own deficiencies into the question -- and ignored the one about invariants
Well the Liskov question was much more trivia than invariants. Liskov mere formalized in an academic sense what many skilled programmers already knew, so many reading Liskov's abstract would probably take a pass. Do not confuse the concept with the name of someone who wrote about the concept.
What may seem new and innovative to the trade to a person with modest experience may not seem new to someone with extensive experience. Agile for instance, a bit of it seemed merely common sense to those of us with a bit of experience.
These are concepts. We have a language that we use to discuss software engineering concepts.
Correct on the first, wrong on the second with respect to Liskov, you have a niche buzzword. Many SOLID adherent may not recall the name that L represents but still recall the concept.
Focus on a buzzword recognition and you may miss someone who actually understands the concept very well. Again, I refer back to your first sentences which were absolutely spot on correct: "A perfectly fine answer is "No, I've never heard of it before. What is it?" Then we can have a conversation."
That is a very poor question. It exhibits the common naive programming test that is more of a trivia test, or basically a programming test derived from some old college quiz. Better questions ask people to perform relevant tasks or to discuss relevant problems. For example rather than ask about "Liskov", ask about the concepts involved:
Q. A Square class is derived from a Rectangle class. Function foo has a non-const parameter of type Rectangle. Function foo is called with using a variable of type Square. Is this safe, if not why?
Note that in the discussion they may inadvertently discuss class invariants.
You say that some programmers do not understand or lack certain tools. The same is true for managers, and your hypothetical questions suggest that you may be lacking the tools necessary to conduct technical interviews and evaluations.
It looks like its actually an underlying issue with Chromium, which is what powers Electron, the UI framework which VS Code is based on. Simple CSS Keyframe Animation Causes Too High CPU Usage
Well if this is the development approach for modern tools and applications I'm glad I splurged and upgraded to the i7 during build to order. It does seem true that software development practices "grow" to fit whatever amount of CPU resources are available.
Is it rendering the cursor specifically at 60 FPS, or is it the entire active window? Because I can imagine a good reason for rendering the active window in an IDE every frame.
Pro tip: when rendering an entire window, or an entire screen in a game, you might want to consider "dirty rectangles" and only redraw what has changed.
You are apparently unaware of how long it takes a regulatory agency to pass a regulation, and that this timeframe is MANDATED in law. The FCC could not pass this regulation in anything short of YEARS without violating the law. My bet is they started working on it in year 4 and it made it through the process 3.5 years later. See by Law to even start work on this there are mandatory publication and comment periods that are typically 6 months to a year, often several of them and delays are included before they can do each step. Comments like yours are perfect examples of the ignorance of how government works.
Actually you prove my point. In your scenario the Obama administration did nothing for 4 years and timed things so that the new rule would not be authorized until the end of his term. Don't want to affect those re-election contributions.
Plus his FCC could have had an implementation plan ready to go once the authority was given, as you suggest, they had years to come up with such a plan. Don't be naive, it was left for the new administration for a reason.
Yet another freedom evaporates thanks to corporate greed and political corruption
Actually this "freedom" never made it out of the vapor stage. The power to enact the policy was given last October, it wasn't to be actually implemented until around now. No "freedom" has been lost because no such "freedom" had ever been acquired. We are still operating under the same rules Obama thought were just fine during his 8 years, when he and his party needed corporate donations. And as if Hillary wouldn't have drafted the policy in some weak manner with workarounds for corporations so her corporate donations could keep rolling in.
The 'socialists' gave the FCC the power to prevent this, asshole.
Gave the power 5 months ago, was it even implemented yet? Apparently for nearly 8 years it was not important. In reality it was just a "mine" planted for political games, a manufactured talking point.
Also in reality this "socialist" president was quite fond of surveillance, drones, extrajudicial killings of US citizens, etc.
You're still eligible for the same severance package - it's the law. You cannot be denied it because you refused to do a job you were not hired for and never agreed to as part of your job description.
No, these severance packages in the US that I am referring to are not required by law. They are something separate from the benefits stated in your original employment agreement. They are a new additional contract stating if you do this, we will give you that. "This" can involve assistance with your replacement's training.
Also, they have to give you a reference no matter what. Most employers nowadays will, as a matter of policy, limit it to "yes, we can confirm that Jane Smith worked here from YYYY.MM.DD to YYYY.MM.DD" because they don't want lawsuits. Get your personal references from co-workers, not your bosses.
Prospective employers can contact you previous employers, not just your stated references. Some companies have policies limiting responses to confirming date, title, etc, not "many". Other factual responses are allowed at some companies too, for example is the person eligible for re-hire. Furthermore if you are laid off it is not uncommon to get a letter of reference. Absence of such a letter can be suspicious. And "would you provide a letter of reference for [name]" is another fact based question that may be answered without prompting a lawsuit.
I watched an interesting lecture on Youtube the other day comparing the tank manufacturing strategies of the US, Germany and Russia during WWII. It turns out that all three were different: the US used efficient assembly lines and precise tooling to mass-produce standardized tanks of mid-range cost/complexity, the Russians zerg-rushed low-quality tanks using massive amounts of cheap labor and simple tooling, and the Germans used skilled craftsmen to build high-quality tanks and constantly improved the design (so that each tank was pretty close to unique). Guess which strategy was least successful...
The strategy with the weakest anti-armor airpower.;-)
Soviet success on the ground had a lot to do with the IL-2 Sturmovik close support aircraft. Plus there are the anti-tank guns on the ground. Tank v tank is only a part of the complex and deadly environment tankers tried to survive.
In Generation Kill there was a scene where some marines were dressed down for shooting a technical with a TOW. I think the cost/benefit of munitions and targets of opportunity was one of the reasons that cannons got put back on ground strike aircraft.
For the Marines its not necessarily the cost itself but the number of TOWs available to them. The Marines are not as well funded as the Army or Navy and are adept at improvisation and making do for very good reasons. The Marines will probably purchase some long barreled high choke skeet/trap shotguns to deal with small drones. As they purchased civilian hunting rifles during Viet Nam as necessary to equip snipers and designated marksmen. Its not that such things won't happen in the Army as well, its that in the Army these ideas will largely be confined to the less senior officers. The more senior officers more likely to turn to some research project with a hollywood looking solution.
When were cannons taken off ground strike aircraft? Fighters yes, that was a bad idea of the 1960s. I believe exterior pod mounted cannons were developed for these fighters for when they were to be used in ground attack roles. And these were hastily repurposed for air-to-air once the error of a pure missile fighter became apparent.
Question Who won WW2? Britain and France who lost their colonies and became second rate powers or Germany who got out of sanctions and became an economic powerhouse. Winning but spending too much to win is no win at all
Germany lost 10% of its population to the war, about 7 million people, that is a cost to be factored in as well. The UK's casualties were 6.4% of Germany's, about 450 thousand, and Frances were 8.5%, about 600 thousand. The cost of war includes blood not just gold.
Plus there was half the country becoming a vassal state of the Soviet Union for nearly 50 years. That is losing your own country, not a colony.
Bring able to make individual choices is the basis of feminism, you prat. Including modern feminism.
In theory. In practice some choices are more "correct" than others. Never underestimate the ability of politicians and their minions to screw up a good idea.
Remove any barriers but let them choose. Maybe some fields are not inherently interesting, we have evolved to have different capabilities and perspectives. If this results in preferences so be it. Let people do what they prefer.
Perhaps they believe they won't succeed in some fields so they don't try. In that case, their barrier is their own prejudice.
That seems an issue of preparation, of introduction to the field. That's the sort of barrier I would remove. Recall "shop" classes in high school? Similar thing, everyone takes a required "intro to programming" type shop class. For those that happen to be somewhat interested they can take the elective more advanced version of that shop class. Not unlike the successful model of decades past.
Instead of a reboot, I'd like to see John Steakley's 'Armor', which basically borrows the premise of Starship Troopers the book (a bug war on an alien planet) but is primarily action focused. When I saw the first preview for the ST movie, I actually thought it *was* Armor before they showed the title. Also, it's just a really good book.
Lets complete the "trilogy" and get a Forever War as well.
And then for the hell of it make an Old Man's War.
Don't women get a say? Must they be 50/51% of every field? Maybe pushing women towards a particular field is no better than pushing them away from a particular field. Remove any barriers but let them choose. Maybe some fields are not inherently interesting, we have evolved to have different capabilities and perspectives. If this results in preferences so be it. Let people do what they prefer.
Never in my wildest dreams did I imagine they'd take the route of Neo being future-Jesus who has completely unexplainable powers in the real world, and we're going to abandon any aspirations of science fiction and go headlong into pure fantasy.
Any sufficiently advanced science/technology is indistinguishable from magic.;-)
Slashdot Mirror
To be clear, by knowledge I do not mean regurgitating textbook definitions. What I want demonstrated is a practical usage of some bit of knowledge. [Borrowing from another post] Rather than ask:
Q. What is the Liskov substitution principle?
I would probably ask something like:
Q. A Square class is derived from a Rectangle class. Function foo has a non-const parameter of type Rectangle. Function foo is called using a variable of type Square. Is this safe, if not why?
Opening VIM is the easy part. If you can successfully CLOSE VIM afterwards, that's the passing point.
Extra credit for a minimal number of beeps while using it. :-)
You ask me what a Liskov substitution is in an interview as a serious primary question and not just as trivia, I'm going to give you a deer in headlights. Afterwards, I'm going to look it up, find out what it was, and even if your company gave me a job offer I'm going to reject it because there's no way in hell I want to work with people who would ask obscure trivia as serious job interview questions. I would never fault a person for not knowing what Levenshtein distance was off the top of their heads.
After a lead engineer gave me a written programming test with a lot of such trivia questions I was back in the project managers office. He asked me how things went. I said OK, I skipped some questions since they were trivia easily looked up in references, things I have not needed to memorize since taking an exam in college. I then went on to describe why their test was flawed and how a better test would go about things. His response, "you're right, the test does sound like something just slapped together in a half assed fashion." I received an offer a few days later.
A perfectly fine answer is "No, I've never heard of it before. What is it?" Then we can have a conversation.
Perfect. Sadly, then you write the following ...
I'm not looking for trivia experts ... A lot of my questions are to probe whether they have bothered to read books that are not typically part of the college curriculum but are important references for a professional software engineer.
Perhaps a beginning software engineer. As one with 30+ years of experience I might look at a summary of Martin's SOLID topic and think "not much new here", low priority or pass.
"S Single responsibility principle
a class should have only a single responsibility (i.e. only one potential change in the software's specification should be able to affect the specification of the class)
O Open/closed principle
“software entities should be open for extension, but closed for modification.”
L Liskov substitution principle
“objects in a program should be replaceable with instances of their subtypes without altering the correctness of that program.” See also design by contract.
I Interface segregation principle
“many client-specific interfaces are better than one general-purpose interface.”
D Dependency inversion principle
one should “depend upon abstractions, [not] concretions.”"
https://en.wikipedia.org/wiki/...
I find it interesting that so many latched on the Liskov question -- and read their own deficiencies into the question -- and ignored the one about invariants
Well the Liskov question was much more trivia than invariants. Liskov mere formalized in an academic sense what many skilled programmers already knew, so many reading Liskov's abstract would probably take a pass. Do not confuse the concept with the name of someone who wrote about the concept.
What may seem new and innovative to the trade to a person with modest experience may not seem new to someone with extensive experience. Agile for instance, a bit of it seemed merely common sense to those of us with a bit of experience.
These are concepts. We have a language that we use to discuss software engineering concepts.
Correct on the first, wrong on the second with respect to Liskov, you have a niche buzzword. Many SOLID adherent may not recall the name that L represents but still recall the concept.
Focus on a buzzword recognition and you may miss someone who actually understands the concept very well. Again, I refer back to your first sentences which were absolutely spot on correct: "A perfectly fine answer is "No, I've never heard of it before. What is it?" Then we can have a conversation."
What is the Liskov substitution principle?
That is a very poor question. It exhibits the common naive programming test that is more of a trivia test, or basically a programming test derived from some old college quiz. Better questions ask people to perform relevant tasks or to discuss relevant problems. For example rather than ask about "Liskov", ask about the concepts involved:
Q. A Square class is derived from a Rectangle class. Function foo has a non-const parameter of type Rectangle. Function foo is called with using a variable of type Square. Is this safe, if not why?
Note that in the discussion they may inadvertently discuss class invariants.
You say that some programmers do not understand or lack certain tools. The same is true for managers, and your hypothetical questions suggest that you may be lacking the tools necessary to conduct technical interviews and evaluations.
If coding tests were so important they would require you to submit to them every year.
Your job performance is the followup "coding test".
They are in fact nothing to do with code but giving someone a problem and watching how they react.
If they react in a catastrophically bad manner that is useful data.
No test can in fact reveal how much knowledge a person has, just how well they do in certain situations.
True, but it can help demonstrate how little they know, if they fail one situation after another after another.
It looks like its actually an underlying issue with Chromium, which is what powers Electron, the UI framework which VS Code is based on.
Simple CSS Keyframe Animation Causes Too High CPU Usage
Well if this is the development approach for modern tools and applications I'm glad I splurged and upgraded to the i7 during build to order. It does seem true that software development practices "grow" to fit whatever amount of CPU resources are available.
Is it rendering the cursor specifically at 60 FPS, or is it the entire active window? Because I can imagine a good reason for rendering the active window in an IDE every frame.
Pro tip: when rendering an entire window, or an entire screen in a game, you might want to consider "dirty rectangles" and only redraw what has changed.
13 per cent CPU. For a blinking cursor. That's... impressive. But not in a good way
This is why coding tests are given during job interviews.
13% of CPU at 60 fps. Maybe they shouldn't use Javascript and a cross platform framework for drawing the cursor. ;-)
You are apparently unaware of how long it takes a regulatory agency to pass a regulation, and that this timeframe is MANDATED in law. The FCC could not pass this regulation in anything short of YEARS without violating the law. My bet is they started working on it in year 4 and it made it through the process 3.5 years later. See by Law to even start work on this there are mandatory publication and comment periods that are typically 6 months to a year, often several of them and delays are included before they can do each step. Comments like yours are perfect examples of the ignorance of how government works.
Actually you prove my point. In your scenario the Obama administration did nothing for 4 years and timed things so that the new rule would not be authorized until the end of his term. Don't want to affect those re-election contributions.
Plus his FCC could have had an implementation plan ready to go once the authority was given, as you suggest, they had years to come up with such a plan. Don't be naive, it was left for the new administration for a reason.
Yet another freedom evaporates thanks to corporate greed and political corruption
Actually this "freedom" never made it out of the vapor stage. The power to enact the policy was given last October, it wasn't to be actually implemented until around now. No "freedom" has been lost because no such "freedom" had ever been acquired. We are still operating under the same rules Obama thought were just fine during his 8 years, when he and his party needed corporate donations. And as if Hillary wouldn't have drafted the policy in some weak manner with workarounds for corporations so her corporate donations could keep rolling in.
The 'socialists' gave the FCC the power to prevent this, asshole.
Gave the power 5 months ago, was it even implemented yet? Apparently for nearly 8 years it was not important. In reality it was just a "mine" planted for political games, a manufactured talking point.
Also in reality this "socialist" president was quite fond of surveillance, drones, extrajudicial killings of US citizens, etc.
You're still eligible for the same severance package - it's the law. You cannot be denied it because you refused to do a job you were not hired for and never agreed to as part of your job description.
No, these severance packages in the US that I am referring to are not required by law. They are something separate from the benefits stated in your original employment agreement. They are a new additional contract stating if you do this, we will give you that. "This" can involve assistance with your replacement's training.
Also, they have to give you a reference no matter what. Most employers nowadays will, as a matter of policy, limit it to "yes, we can confirm that Jane Smith worked here from YYYY.MM.DD to YYYY.MM.DD" because they don't want lawsuits. Get your personal references from co-workers, not your bosses.
Prospective employers can contact you previous employers, not just your stated references. Some companies have policies limiting responses to confirming date, title, etc, not "many". Other factual responses are allowed at some companies too, for example is the person eligible for re-hire. Furthermore if you are laid off it is not uncommon to get a letter of reference. Absence of such a letter can be suspicious. And "would you provide a letter of reference for [name]" is another fact based question that may be answered without prompting a lawsuit.
I watched an interesting lecture on Youtube the other day comparing the tank manufacturing strategies of the US, Germany and Russia during WWII. It turns out that all three were different: the US used efficient assembly lines and precise tooling to mass-produce standardized tanks of mid-range cost/complexity, the Russians zerg-rushed low-quality tanks using massive amounts of cheap labor and simple tooling, and the Germans used skilled craftsmen to build high-quality tanks and constantly improved the design (so that each tank was pretty close to unique). Guess which strategy was least successful...
The strategy with the weakest anti-armor airpower. ;-)
Soviet success on the ground had a lot to do with the IL-2 Sturmovik close support aircraft. Plus there are the anti-tank guns on the ground. Tank v tank is only a part of the complex and deadly environment tankers tried to survive.
In Generation Kill there was a scene where some marines were dressed down for shooting a technical with a TOW. I think the cost/benefit of munitions and targets of opportunity was one of the reasons that cannons got put back on ground strike aircraft.
For the Marines its not necessarily the cost itself but the number of TOWs available to them. The Marines are not as well funded as the Army or Navy and are adept at improvisation and making do for very good reasons. The Marines will probably purchase some long barreled high choke skeet/trap shotguns to deal with small drones. As they purchased civilian hunting rifles during Viet Nam as necessary to equip snipers and designated marksmen. Its not that such things won't happen in the Army as well, its that in the Army these ideas will largely be confined to the less senior officers. The more senior officers more likely to turn to some research project with a hollywood looking solution.
When were cannons taken off ground strike aircraft? Fighters yes, that was a bad idea of the 1960s. I believe exterior pod mounted cannons were developed for these fighters for when they were to be used in ground attack roles. And these were hastily repurposed for air-to-air once the error of a pure missile fighter became apparent.
Question Who won WW2? Britain and France who lost their colonies and became second rate powers or Germany who got out of sanctions and became an economic powerhouse. Winning but spending too much to win is no win at all
Germany lost 10% of its population to the war, about 7 million people, that is a cost to be factored in as well. The UK's casualties were 6.4% of Germany's, about 450 thousand, and Frances were 8.5%, about 600 thousand. The cost of war includes blood not just gold.
Plus there was half the country becoming a vassal state of the Soviet Union for nearly 50 years. That is losing your own country, not a colony.
Watch the sequels. They make the first look better. :-)
Bring able to make individual choices is the basis of feminism, you prat. Including modern feminism.
In theory. In practice some choices are more "correct" than others. Never underestimate the ability of politicians and their minions to screw up a good idea.
as a tangent to that Another Scalzi title, Redshirts is in Development purgatory/limbo
Now there's a great example of combining "borrowing" and "originality". ;-)
Remove any barriers but let them choose. Maybe some fields are not inherently interesting, we have evolved to have different capabilities and perspectives. If this results in preferences so be it. Let people do what they prefer.
Perhaps they believe they won't succeed in some fields so they don't try. In that case, their barrier is their own prejudice.
That seems an issue of preparation, of introduction to the field. That's the sort of barrier I would remove. Recall "shop" classes in high school? Similar thing, everyone takes a required "intro to programming" type shop class. For those that happen to be somewhat interested they can take the elective more advanced version of that shop class. Not unlike the successful model of decades past.
Instead of a reboot, I'd like to see John Steakley's 'Armor', which basically borrows the premise of Starship Troopers the book (a bug war on an alien planet) but is primarily action focused. When I saw the first preview for the ST movie, I actually thought it *was* Armor before they showed the title. Also, it's just a really good book.
Lets complete the "trilogy" and get a Forever War as well.
And then for the hell of it make an Old Man's War.
Don't women get a say? Must they be 50/51% of every field? Maybe pushing women towards a particular field is no better than pushing them away from a particular field. Remove any barriers but let them choose. Maybe some fields are not inherently interesting, we have evolved to have different capabilities and perspectives. If this results in preferences so be it. Let people do what they prefer.
"God always makes a better idiot" was the one my cs college professors always used
The "idiot" is just an instance of using a Monte Carlo method search of solution space. :-)
Never in my wildest dreams did I imagine they'd take the route of Neo being future-Jesus who has completely unexplainable powers in the real world, and we're going to abandon any aspirations of science fiction and go headlong into pure fantasy.
Any sufficiently advanced science/technology is indistinguishable from magic. ;-)