Related
This might be a naive question, but i am really interested to know why logic was developed to be used in AI. In particular, what was the need to develop first order logic and PDDL in AI, if we could do the programming using simple atomic representation of states? Again, I realize this is a really basic question!!
So your question is about: why do we program/model on a first-order level instead of a propositional level? Simply because it is more concise.
You can make propositions like "All humans can think." with a first-order language and don't have to state "Alice can think. Bob can think. Carol can think. ...".
If you look at some PDDL planning problems from the IPC, there are sometimes ground versions that are formulated on a propositional level. And the files are much larger. You don't want to write those by hand.
I don't know about PDDL, but first order logic was developed before computers ever were invented, so it wasn't for use in AI. It tells you what arguments are valid.
I'm looking at the requirements for automated software verification, i.e. a program that takes in code (ordinary procedural code written in languages like C and Java), generates a bunch of theorems saying that each loop must eventually halt, no assertion will be violated, there will never be a dereference of a null pointer etc., then passes same to a theorem prover to prove they are actually true (or else find a counterexample indicating a bug in the code).
The question is what kind of logic to use. The two major positions seem to be:
First-order logic is just fine.
First-order logic isn't expressive enough, you need higher order logic.
Problem is, there seems to be a lot of support for both positions. So which one is right? If it's the second one, are there any available examples of things you want to do, that verifiers based on first-order logic have trouble with?
You can do everything you need in FOL, but it's a lot of extra work - a LOT! Most existing systems were developed by academics / people with not a lot of time, so they are tempted to take short cuts to save time / effort, and thus are attracted to HOLs, functional languages, etc. However, if you want to build a system that is to be used by hundreds of thousands of people, rather than merely hundreds, we believe that FOL is the way to go because it is far more accessible to a wider audience. There's just no substitute for doing the work; we've been at this for 25 years now! Please take a look at our project (http://www.manmademinions.com)
Regards, Aaron.
In my practical experience, it seems to be "1. First-order logic is just fine". For examples of complete specifications for various functions written entirely in a specification language based on first-order logic, see for instance ACSL by Example or this case study.
First-order logic has automated provers (not proof assistants) that have been refined over the years to handle well properties that come from program verification. Notable automated provers for these uses are for instance Simplify, Z3, and Alt-ergo. If these provers fail and there is no obvious lemma/assertion you can add to help them, you still have the recourse of starting up a proof assistant for the difficult proof obligations. If you use HOL on the other hand, you cannot use Simplify, Z3 or Alt-ergo at all, and while I have heard of automated provers for high-order logic, I have never heard them praised for their efficiency when it comes to properties from programs.
We've found that FOL is fine for most verification conditions, but higher order logic is invaluable for a small number, for example for proving properties about summation of the elements in a collection. So our theorem prover (used in Perfect Developer and Escher C Verifier) is basically first order, but with the ability to do some higher order reasoning as well.
Closed. This question is off-topic. It is not currently accepting answers.
Want to improve this question? Update the question so it's on-topic for Stack Overflow.
Closed 10 years ago.
Improve this question
Does anyone actually think this is a good reason to "Dumb down" your code?
When a manager asks you to make your code simple (in terms of technology skills required to understand it) at the cost of more verbose cluttered code what should you do?
I highly disagree. Junior developers will end up being Senior developers. How? By encountering advanced topics that aren't taught in school.
My code base now makes heavy use of Inversion of Control containers. I would never revert my code to the old way because a junior developer had issues groking IoC. Instead I would take them out for a beer after work and discuss it. The more the junior dev learns the less hand holding needs to be done.
Here's a blog post discussing this very topic.
If you're constantly dumbing down your code or designs, it's a pretty good way to make sure your junior developers stay dumb. Challenge them and use it as a mentoring opportunity. Of course, some will never learn, but you've got bigger problems at that point.
It's not just pointy-haired bosses either. As a senior dev, it's often difficult to resist the urge to mommy junior developers. "Oh I'll just do this part because it's way too hard for them", or it'll take them too long, or they'll get way off in the weeds.
And finally, make sure you strike a balance between idiomatic code that uses the full power of a language vs idiomatic code that abuses that power. There's no reason you need to override the || operator just to run its args in two separate threads. At least dumb the code down a little for your older, dumber, future self.
Well, I think it's reasonable to avoid using "clever" language constructs unless they really, really make the code better - at which point if a junior developer sees it, hopefully they'd ask rather than just being flumoxed.
Here's an alternative way of phrasing it though: "Write your code so that it's easy enough to understand that if you get called at 3am and asked to fix a bug in it, you can still understand it."
Seriously, make it as easy to understand as possible. That doesn't mean a comment every other line - it means a comment where the purpose of a piece of code isn't obvious, and only then where the preferred choice of "well make it obvious then" doesn't work.
There's a difference between puzzle code and complex code.
I've found that the single biggest issue is that there is a big difference between "easy to understand by reading" versus "well-factored", and that the two goals are often in direct tension with one another. In well-factored code, there is a lot more jumping around between classes and a lot of virtual dispatch, so the path through the code is very non-linear.
Yes readability and being able to easily understand code is a big part of maintainability in my opinion.
Well if you intend to maintain your code forever, never change jobs, never feel the urge to work on something new, and can assure everyone you will never be hit by a truck, then sure there is no need to dumb down that puzzle code.
No. In the past, I've learned a lot from seeing the tricks of more experienced developers. I'd much rather have had the opportunity to learn something new from them than have had them dumb things down for me.
Its a balancing act...
If any 3 people on your team can 'read' your code and know what its doing... no need to change. However if you're the only person who can understand your code (no matter how rad/clever you think it is).. maybe you should take it down a few notches.
Another guideline to help would be to 'Try the simplest thing that works.' All the latest buzz words are nice to know however what it is even more important is having the skill to spot where you could get by without using them. You don't need to spray paint your code with IOC or Frameworks or Design Patterns...
The manager's side of this argument is sorely missed in this thread :) (and for the record.. I'm not one). His/Her major concern being he doesn't want a dark area of code that no one else dares to venture into.. so if you can convince your boss that a few other people on the team can make an arbitrary fix (or better yet.. show an actual bug fixed by someone else) - the mgr should let you off the hook. Disagreeing with your boss is another art :).. but you can talk things out usually.
You dont have to go all the way backward to Lowest Common Denominator.. strike a balance.
Your goal should not be for your code to be easy to understand for a junior developer. Instead, it should be easy to understand for a maintainence programmer.
This means:
Local "complexity" is okay, when needed. If they see the complex code they'll know they need to dig deeper.
Hidden complexity is bad. If you can't see that changing a piece of code will have subtle side effects then maintaining the code will be a nightmare.
New technologies that are visible are also okay, when not taken to extremes.
This is because those that maintain code rarely have the same overall understanding of the system. Or the time to develop it.
I disagree with the manager: What needs to be simple is the code, not the technology used to write it.
I would, however, impose a closely related requirement:
The internal documentation states clearly what technologies are needed to understand this code, and it gives references to places where those technologies can be learned.
For example, even as a senior developer, I find all matrix codes baffling. But if somebody gives me a reference to the right part of Numerical Recipes, I can puzzle out the details.
Yes. It's a very valid reason to take it down a notch. The reality is that a very, very large number of developers (as in most) are at the junior level.
As far as what you should do... Say "Yes Sir" or "Yes Ma'am" and do it. There is only one boss in that relationship.
UPDATE:
As some people seem to think that having a jr dev learning advanced topics while wading through obfuscated code I want to throw one more thing in here.
When ANY developer (jr or otherwise) runs into code they don’t understand, their first approach is to refactor it into something that is understandable. This is called the “Wow that code is crap I must rewrite it!” syndrome. I’m willing to bet everyone on this board has experienced it. So, as a business owner, do I want to pay for code to be developed each time a new person comes by or do I want to pay for new features to be added?
Guess which person I’m going to keep around longer.
If you dumb down your code, you're going to be stuck working with dummy junior programmers who will never be familiar with advanced coding techniques. If there's any verbose code that's trying to express an inherently complex procedure that you wrote, the aforementioned junior developer probably wouldn't be able to see the forest for the trees anyways. And they'd probably screw up if they had to express a complex concept if all they knew were basic primitive constructs whereas if they knew how to express what they meant tersely and elegantly, the code has a better chance of being correct.
Scott Muc said:
"I've found that the single biggest issue is that there is a big difference between "easy to understand by reading" versus "well-factored", and that the two goals are often in direct tension with one another. In well-factored code, there is a lot more jumping around between classes and a lot of virtual dispatch, so the path through the code is very non-linear."
Quoted for truth, and I think this is one of the biggest problems with C++ code in general. If you're the one that wrote the code, it's pretty easy to come up with a very complicated set of stuff that is well factored, makes lots of sense if you already know it, works well, and generally resembles a diamond crystal, etc. but which, from the perspective of someone who's trying to figure out how you got there and why things are the way they are and how things work, and how one might make changes that fit into the existing system and satisfy new requirements, is almost completely opaque and impenetrable.
How does this kind of situation help maintainability? That situation is one of my main beefs with C++ programmers. Far better to have a mess of plain C code which can be hacked upon than a diamond crystal of inpenetrably super-factored code which nearly nobody can figure out how to sensibly modify without smashing the crystalline structure.
One way to "dumb down" code that I actually think is an excellent practice is to use longer variable names and longer function names. Naming variables and functions to make their purpose easily understandable is a significant engineering task, IMHO, and takes extra effort on the part of the original author of the code. Damian Conway has some excellent examples in "Perl Best Practices". Some examples include: Prefer "final_total" to "sum"; prefer "previous_appointment" to "previous_elem", prefer "next_client" to "next_elem". Prefer "sales_records" to "data". Etc. He also pushes for using grammatical templates (Noun-adjective) and staying consistent. Don't use max_displacement one place and then use VelocityMax in another. Index variables need real names too:
sales_record[i] vs sales_record[cancelled_transaction_number]
I frequently "refactor" my code at the end of a development cycle by finding new names for all my functions and variables. In a modern editor it's trivial to change them all, and it's only at the end that I really figure out what I used them for. "Dumbing down" code this way isn't classic C, but it's easier for me when I come back months later asking WTF did I do?
It depends on the code. Is this something being shipped in your flagship product that requires use of the features your manager wants you to remove for performance reasons? If the answer is yes I would try to have your manager let you keep the code and just write up a document explaining in detail the section of code that is hard to understand. If it's an internal app that needs to be maintained by lots of different people and the complex features can be removed with out negatively affecting the program remove them and pick more important battles to fight.
You should just remind your boss that you can build rocket ships or chicken coops, and he will have to pay you the same for either one. Do what they say but generally an environment like that lends itself to people looking for a new environment.
The old quote is appropriate here:
Make everything as simple as possible,
but not simpler.
I've known developers who wrote highly obfuscated code that they felt was advanced but which the rest of the team felt was unreadable and unmaintainable. Part of this involved overuse of advanced language features (in C, the ternary operator and the comma operator) and writing in an obscure personal idiom (for example, replacing ptr->item with (*ptr).item everywhere) that no-one else would ever be able to maintain. The author was trying to outsmart the optimizer (which to be fair, was far from good).
Note: I'm not suggesting that "x = (p == NULL) ? "default" : p->value;" is complicated, but when someone uses ternary operators that span many lines, are nested, and make heavy use of the comma operator, code quickly becomes unreadable.
In this sort of case, "dumbing down" the code would have been a good idea. The problem was not advanced algorithms nor advanced language features, but overuse and inappropriate use of advanced language features, and an obscure personal idiom.
However, in the case you are asking about, where the manager's changes make the code more difficult to read and maintain, I agree with you and the others who have responded. I just wanted to point out the alternative that no-one else has mentioned.
I suggest keeping the code in a "Geeky-level" and comment it well so that the juniors can understand the intention behind the code and simultaneously learn a better way (or a right way) to code, so we have the best of both he worlds.
I think it is the manager's way of politely telling you that your code is too obfuscated/complex/jumbled/puzzle code...whatever you want to call it. Sometimes we get so involved writing our codes that we forget that someone else will have to come along and read it later.
I learned it the hard way and, in retrospect, find that it was the better way. Let the cycle repeat itself.
I agree 100% with the argument. With one major addition: Train your junior developers until they understand your code ;-)
I'm talking about using "unusual" technologies. In this case it's JQuery.
This issue came up when I was writing a wizard control for user registration.
The navigation menu needed to be customised and the current step in the wizard had to have a different css class in the menu. This meant I needed to get access to the currently selected step when generating the menu. My solution was to output the current step index in a hidden html field which could then be read by JQuery in order to customise the css.
I thought that this was much nicer and cleaner than using the databinding syntax in ASP.NET which doesn't have compile-time checking and messes up the layout of the html.
The databinding solutions is "standard" while the JQuery one is "unusual" which means that it's less likely to be understood by a junior.
I'm trying more and more these days to provide the required data for the UI rather than hack it into the UI with databinding which is why I added the hidden field with the current step index.
It is simply impossible to make progress or to innovate in any industry without doing things that others don't understand. Innovation is necessarily blasphemous. Why? Because if you're doing things that make sense to everyone else around you, the odds are you're not the first one doing it. ;)
That being said, there is a significant difference between doing something that is difficult to understand simply because it's a new or complicated problem versus doing something that's difficult to understand because you're trying to show off or you think confusing people will somehow gain you job security (which I've never seen work, but I've heard of people trying).
Should you make things easy to understand? Yes absolutely, as much as humanly possible. However a program that works and does its job well is the higher priority.
The manager's complaint should never be "don't do this because our junior guys don't understand it" -- it should only ever be "do x instead of y whenever feasible because x is easier to read / understand". This also assumes that x and y are equivalent (accept the same input and produce the same result).
I can't stand when managers do that... I've had three different managers bawl me out for using perfectly normal code the way it was designed to work, not because I was doing anything complicated, but rather only because they felt like it was too much effort for the other guys on our team to go RTFM on the language we were using. As a management strategy, that's totally backwards. It's like being the Holy Roman Catholic church and insisting that the laymen are too dumb to be trusted with literacy.
If you want to know really how ridiculous some of these managers get, try this: I had one manager bawl me out for declaring a variable as a type of "boolean" because he didn't feel the other programmers could handle it. Actually when I asked why, his answer was "because we don't do that here", which is a non-answer, but I interpreted it to mean "dumb it down". They were also berating me for that and similar practices as though it should be obvious that good programming habits were actually "bad" and that I should already know why even though they had never expressed a preferred programming style (either formally or informally). Needless to say, it was a bad job.
Make sure you can understand what it does 6 months down the road.
When in doubt, COMMENT your code. That's what comments are for.
One of my personal programming demons has always been complex logic that needs to be controlled by if statements (or similiar). Not always necessarily that complex either, sometimes just a few states that needs to be accounted for.
Are there any tools or steps a developer can perform during design time to help see the 'states' and take measures to refactor the code down to simplify the resulting code? I'm thinking drawing up a matrix or something along those lines...?
I'd recommend a basic course in propositional logic for every aspiring programmer. At first, the notation and Greek letters may seem off-putting to the math-averse, but it is really one of the most powerful (and oft-neglected) tools in your skillset, and rather simple, at the core.
The basic operators, de Morgan's and other basic laws, truth tables, and existence of e.g. disjunctive and conjunctive normal forms were an eye-opener to me. Before I learned about them, conditional expressions felt like dangerous beasts. Ever since, I know that I can whip them into submission whenever necessary by breaking out the heavy artillery!
Truth tables are basically the exhaustive approach and will (hopefully) highlight all the possibilities.
You might like to take a look at Microsoft Pex, which can be helpful for spotting the fringe cases you hadn't thought of.
I think that the developer is asking how to make his life easier when dealing with complex if code.
The way that I handle complex if code is to code as flat as possible and weed out all negations first. If you can get rid of compound if by placing a portion of it above, then do that.
The beauty of simplicity is that it doesn't take a book or a class to learn it. If you can break it up, do so. If you can remove any part of it, do so. If you don't understand it, do it differently. And flat is almost always better than nested (thanks python!).
It's simpler to read:
if(broken){
return false;
}
if (simple){
doit();
return true;
}
if(complicated){
divide();
conquor();
}
if(extra){
extra();
}
than it is to read:
if(!broken && (simple || complicated)){
....
}
return false;
Truth tables and unit tests - draw up the tables (n dimensional for n variables), and then use these as inputs to your unit test, which can test each combination of variables and verify the results.
The biggest problem I've seen through the years with complex IFs is that people don't test all the branches. Make sure to write a test for each possible branch no matter how unlikely it seems that you will hit it.
You might also want to try Karnaugh maps, which are good for up to 4 variables.
If you haven't already, I'd highly suggest reading Code Complete. It has a lot of advice on topics such as this. I don't have my copy handy at the moment, otherwise I'd post a summary of this section in the book.
Split the logic down into discrete units (a && b, etc.), each with their own variable. Then build these up using the logic you need. Name each variable with something appropriate, so that your complex statement is fairly readable (although it may take up several extra lines and a fair few temporary variables).
Any reason you cannot just handle the logic with guard statements?
Karnaugh maps can be nice ways of taking information from a truth table (suggested by Visage) and turning them into compact and/or/not expressions. These are typically taught in an EE digital logic course.
Have you tried a design pattern? You might look into what is known as the Strategy pattern: http://en.wikipedia.org/wiki/Strategy_pattern
Check out the nuclear option: Drools. There's quite a lot to it-- took me a day or two of perusing the literature just to get a handle on its capabilities. But if you have applications where your complex if-then logic is an evolving part of the project (for example, an application with modular algorithms) it might be just the thing.
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 4 years ago.
Improve this question
Many times we find ourselves working on a problem, only to figure out the solution being created is far more complex than the problem requires. Are there controls, best practices, techniques, etc that help you control over complication in your workplace?
Getting someone new to look at it.
In my experience, designing for an overly general case tends to breed too much complexity.
Engineering culture encourages designs that make fewer assumptions about the environment; this is usually a good thing, but some people take it too far. For example, it might be nice if your car design doesn't assume a specific gravitational pull, nobody is actually going to drive your car on the moon, and if they did, it wouldn't work, because there is no oxygen to make the fuel burn.
The difficult part is that the guy who is developed the "works-on-any-planet" design is often regarded as clever, so you may have to work harder to argue that his design is too clever.
Understanding trade-offs, so you can make the decision between good assumptions and bad assumptions, will go a long way into avoiding a needlessly complicated design.
If its too hard to test, your design is too complicated. That's the first metric I use.
Here are some ideas to get design more simpler:
read some programming books and articles, and then apply them in your work and write code
read lots of code (good and bad) written by other people (like Open Source projects) and learn to see what works and what does not
build safety nets (unit tests) to enable experimentations with your code
use version control to enable rollback, if those experimentations take wrong turn
TDD (test driven development) and BDD (behaviour driven development)
change your attitude, ask how you can make it so, that "it simply works" (convention over configuration could help there; or ask how Apple would do it)
practice (like jazz players -- jam with code, try Code Kata)
write same code multiple times, with different languages and after some time has passed
learn new languages with new concepts (if you use static language, learn dynamic one; if you use procedural language, learn functional one; ...) [one language per year is about right]
ask someone to review you code and actively ask how you can make your code simpler and more elegant (and then make it)
get years under your belt by doing above things (time helps active mind)
I create a design etc., and then I look at it and try and remove (agressively) everything that doesn't seem to be needed. If it turns out I need it later when I am polishing the design I add it back in. I do this over several iterations, refining as I go along.
Read "Working Effectively With Legacy Code" by Michael C. Feathers.
The point is, if you have code that works, and you need to change the design, nothing works better than making your code unit testable, and breaking your code into smaller pieces.
Using Test Driven Development and following Robert C. Martin's Three Rules of TDD:
You are not allowed to write any production code unless it is to make a failing unit test pass.
You are not allowed to write any more of a unit test than is sufficient to fail; and compilation failures are failures.
You are not allowed to write any more production code than is sufficient to pass the one failing unit test.
In this way you are not likely to get much code that you don't need. You will always be focused on making one important thing work and won't ever get too far ahead of yourself in terms of complexity.
Test first may help here, but it is not suitable for all situation. And it's not a panacea anyway.
Start small is another great idea. Do you really need to stuff all 10 design patterns into this thing? Try first to do it "stupid way". Doesn't quite cut it? Okay, do it "slightly less stupid way". Etc.
Get it reviewed. As someone else wrote, two pairs of eyes are better. Even better are two brains. Your mate may just see a room for simplification, or a problematic area you thought was fine just because you spend many hours hacking it.
Use lean language. Languages such as Java, or sometimes C++ sometimes seem to encourage nasty, convoluted solutions. Simple things tend to span over multiple lines of code, and you just need to use 3 external libraries and a big framework to manage it all. Consider using Python, Ruby, etc. - if not for your project, then for some private use. It can change your mindset to favor simplicity, and to be assured that simplicity is possible.
Reduce the amount of data you're working with by serialising the task into a series of smaller tasks. Most people can only hold half a dozen (plus or minus) conditions in their head while coding, so make that the unit of implementation. Design for all the tasks you need to accomplish, but then ruthlessly hack the design so that you never have to play with more than half a dozen paths though the module.
This follows from Bendazo's post - simplify until it becomes easy.
It is inevitable once you have been a programmer that this will happen. If you seriously have unestimated the effort or hit a problem where your solution just doesn't work then stop coding and get talking to your project manager. I always like to take the solutions with me to the meeting, problem is A, you can do x which will take 3 days or we can try y which will take 6 days. Don't make the choice yourself.
Talk to other programmers every step of the way. The more eyes there are on the design, the more likely an overcomplicated aspect is revealed early, before it becomes too ossified in the codebase.
Constantly ask yourself how you will use whatever you are currently working on. If the answer is that you're not sure, stop to rethink what you're doing.
I've found it useful to jot down thoughts about how to potentially simplify something I'm currently working on. That way, once I actually have it working, it's easier to go back and refactor or redo as necessary instead of messing with something that's not even functional yet.
This is a delicate balancing act: on the one hand you don't want something that takes too long to design and implement, on the other hand you don't want a hack that isn't complicated enough to deal with next week's problem, or even worse requires rewriting to adapt.
A couple of techniques I find helpful:
If something seems more complex than you would like then never sit down to implement it as soon as you have finished thinking about it. Find something else to do for the rest of the day. Numerous times I end up thinking of a different solution to an early part of the problem that removes a lot of the complexity later on.
In a similar vein have someone else you can bounce ideas off. Make sure you can explain to them why the complexity is justified!
If you are adding complexity because you think it will be justified in the future then try to establish when in the future you will use it. If you can't (realistically) imagine needing the complexity for a year or three then it probably isn't justifiable to pay for it now.
I ask my customers why they need some feature. I try and get to the bottom of their request and identify the problem they are experiencing. This often lends itself to a simpler solution than I (or they) would think of.
Of course, if you know your clients' work habits and what problems they have to tackle, you can understand their problems much better from the get-go. And if you "know them" know them, then you understand their speech better. So, develop a close working relationship with your users. It's step zero of engineering.
Take time to name the concepts of the system well, and find names that are related, this makes the system more familiar. Don't be hesitant to rename concepts, the better the connection to the world you know, the better your brain can work with it.
Ask for opinions from people who get their kicks from clean, simple solutions.
Only implement concepts needed by the current project (a desire for future proofing or generic systems make your design bloated).