Related
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 6 years ago.
Improve this question
What is the best programming language for artificial intelligence purposes?
Mind that using suggested language I must be able to employ any AI technique (or at least most of them).
All the cool bearded gurus in what's left of AI research use Lisp :)
There are two big camps: Common Lisp and Scheme. They have different syntax, etc. Lots of good stuff written for both.
Java is a very popular all-purpose language but a lot of the interesting stuff in AI / Functional Programming, such as passing closures as first-order objects, is clumsy to do in Java.
My personal preference would be to stay away from Windowsy languages like C# and F#. Cool people develop under Unix. Or Linux if they're cool but poor.
Some cool but weird people program in Haskell. A reasonably modern FP language with good performance. I tried it once, it made my brain hurt; but you might be smarter than I am.
UPDATE: Answers to Steve's questions.
I wouldn't be the one paying for a Unix variant; that's what corporations and research institutes do. The idea is, you want to be doing AI research for an outfit that sinks millions into their hardware and doesn't balk at paying a few thousand for an operating system. That's the kind of outfit likely to have good food in the cafeteria and/or pay well for doing fun work. But I'm certainly not knocking Linux.
F# may be cool but I see a whole raft of issues getting it to run on Linux or any other Unix (that's what I meant by "windowsy"), and I don't want to work under Windows (that's what I meant by "personal preference").
To elaborate on the "windowsy" theme: You mention that F# is an OCaml variant. From my own admittedly brief research, it seems that F# is missing functors, OCaml-style objects, polymorphic variants and the camlp4 preprocessor. A functional language without functors? Really? If one were disposed to not like Microsoft, as I admittedly am, one could conclude that they had gone ahead and crowbarred a perfectly good functional language, OCaml, into something they could get to run in their CLR so they could claim to "have" a functional language. Finally, because I don't suspect, I know that Microsoft always prioritizes market dominance over product quality, I don't plan to touch F#. But this is my personal preference, and clearly identified as such, while we're really more concerned with making a good recommendation for mary.ja45 .
I have better reasons to recommend Lisp over F# and even OCaml and Haskell. These are mostly based on the historic preponderance of Lisp over any other language in the AI field.
The bulk of AI literature is based on programs written in Lisp or Prolog. If nothing else, good knowledge of Lisp would allow a student to understand the sample programs. My personal favorite AI megaproject, Cyc, has runtimes in your choice of Common Lisp or C.
In the TIOBE index of programming language (as seen and used in industry), Lisp takes 15th place while Haskell takes 43rd and F# and OCaml place below 50th. Presence on the market correlates with employment opportunities, naturally.
That said, it's quite possible that a number of the younger "AI interesting" languages are poised to skyrocket. If some major research institute published some groundbreaking, defining-the-field research in, say, Scala, you'd see Scala's popularity advance sharply in the research community and, with some lag, in industry.
I (obviously) can't comment on F#'s other qualities but you're as welcome to make recommendations as I was.
Python seems to be used a lot in the general scientific community. It has a lot of libraries available and it's easy to learn.
I'll throw Scala into the pot.
it's usable for functional programming
it can be made as fast as Java
it's a modern language with lot's of nice aspects
Java seems to be a bit popular in AI, too and so you can use all those Java libraries from Scala
I've solved all exercises from a basic AI course in Scala. It worked really well.
If by "all of AI" you also mean machine learning, which I guess, Matlab, R and Python+Scipy should definately be mentioned.
I personally use Clojure for AI programming, and have found it to be a great all-rounder AI language.
Reasons:
It's a Lisp, and Lisps have historically been very strong the the AI field
It's a homoiconic language with powerful macros, so great for code generation and genetic programming. This is a surprisingly useful property for AI programming (and possibly explains some of the success of Lisp in general in this space)
It runs on the JVM and can easily access all the Java libraries for number crunching (Weka, Colt, etc.).
It's good for rapid interactive development - it's very dynamic and you can do pretty much everything interactively in a running Clojure REPL. No need for recompiling etc.
It matters probably whether the programming environment is academic or not, but for most non-academic AI application development I would recommend sticking with a mainstream language like Java or C++. One needs to be able to interface readily with other COTS or open-source software packages, and this can sometimes be difficult or impossible in more "exotic" languages. For academic work this may be a less critical issue.
Additionally, performance can be important for many applications, and mainstream languages generally have the most heavily-optimized compilers, e.g., C++ or Java.
It is true that functional programming languages like LISP, Scheme, etc have specialized features that may make it easier to implement particular AI methods, but I do not believe this to be true for AI-related programming as a whole, e.g., quantitative machine learning methods usually don't require a functional language. If you need access to both functional constructs and general software packages, there are some tools for LISP to help with this, and the recently-developed Clojure is a LISP-variant that runs on the JVM and can access Java libraries. Also, Groovy is another JVM-based language that includes support for closures.
Lastly, some programmers like paradigm flexibility and/or fast prototyping for AI projects. Ruby and Python both see some AI-related usage for this reason as multi-paradigm languages that can also be used for scripting.
Like most things in programming, the best answer for which language to use in AI development will ultimately depend on the needs of your projects.
It really depends on what kind of problem you are looking at. Also, how "deep" you want to go into AI stuff. If you want to learn from the basics and just implement theoretical AI stuff, go with a higher level language-- as in functional programming (and proven in AI) like lisp, or prolog. If you know what problem set you are dealing with and want efficient, go with something like Java, C++ and use a toolkit to do the stuff.
Since you mention Machine Learning look into Weka Toolkit in Java for some of these stuff.
Pick the programming language with AI techniques the same way you pick a language for any other project:
What is the problem you are trying to
solve?
Is there good support available for
the language?
What are the customers requirements?
I would recommend Prolog as a very good programming language used to implement AI systems.
There is no "best" language. Each one has its merits. When I studied AI, mostly we worked with lisp and prolog, but I've been most productive in AI with Java/C# and F# has a lot to offer.
How about a framework written in Java, supporting "High Level Logic" and agent style communication.
http://highlevellogic.blogspot.com/2010/11/when-will-we-have-artificial.html
It also depends on the size of your dataset. For web-scale datasets you may want to use Map-Reduce and that implies Hadoop. Hadoop is in Java -- but you could use any language (Python, etc.) for your Map-Reduce functions.
There is also a java framework called weka, developped by the university of waikato. I don't know wether it anwsers your question, but it may help.
Quoting wikipedia: «Weka supports several standard data mining tasks, more specifically, data preprocessing, clustering, classification, regression, visualization» and more.
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 8 years ago.
Improve this question
since I was at the first year of my University I always envied my fellows (mainly coming from a tech-oriented professional school) for knowing C. I came from a natural-sciences-oriented lyceum and never had programming experience or courses but some summer work with PHP learned from a teach-yourself-PHP-in-7-hours (and my programming interest was very recent). I want to know ... was it a legitimate envy? Does any programmer have to know C? Does C provide deep understanding on how a system works or how programming has to be carried out? I know that when programming C you have to have a strong understanding about buffers, memory, and so on. So I want your opinion on it.
No, you don't have to know C. But knowledge of C (or any other "close to the machine but not assembler" language) greatly enhances your potential as a programmer. Because you will understand a lot more of the inner workings.
And of course knowledge of assembler is also valuable. But in the above paragraph, I wanted to target the low end of computer language. Just because in modern languages, we take so much for granted (OO, extensive libraries, garbage collection to name a few). And yes it helps us programmers to work more efficiently. But it hides some of the machine aspects that we sometimes need to see and that's why it is so important that we need to know the inner workings.
While I think it is important to understand memory allocation, pointers, registers, ... I would say that too much experience in C can also be a barrier to grasp higher level languages, or OO languages. C tends to make you think procedurally and that can be a bad thing in some cases ...
I would definitely not recommend C as a first language, but it can be of great help to do some C at some point...
As a side note, I think that assembly can be much more useful to help you understand basic principle. At the same time assembly is far enough from any language that you would use (unless you work in a very specialized field). That will help you keep a different mindset when doing assembly than when using a higher level language.
This is a quite discussable topic. I personally think that knowing C enhances your ability to work with other languages too when you understand what's going on under the hood. But you don't have to know C to be able to produce high quality code.
Eric Sink has also once thought of that question.
Something I find missing from most of these answers is that C is a very easy language to learn. Everything you'll ever need to know about the syntax is contained in one thin, concise book (K&R), and that includes all the standard libraries. So I'd encourage you to at least skim a book and see what it's all about, even if you don't intend to use it.
That simple C syntax only gets expanded for most modern C-based languages (C++, C#, Java). You can't say you really know those languages until you've mastered at least a subset of the hundreds of libraries that come with them, and that can take months or years of experience.
What's tough about C is that it can expose you to the true nature of the machine underneath. If you really want to grok how a computer works, you need to understand things like pointers, memory allocation, and stack vs. heap vs. executable code. You can learn basic C syntax in a few hours, and that puts you on the road to understanding much more. Saying "I'm a C expert" is just a proxy for saying, "I really understand how a computer works."
You don't necessarily have to know C, but imo every programmer should know about basic machine architecture and how applications interact with the OS and the hardware.
Obviously if you're going to study this, C is a good choice for a language, but not the only option.
Another good reason to know C is that a lot of code is written in C so if you want to learn from others code, it will be very helpful.
In addition to Gamecats answer, in my experience working with people in other languages, there is a difference in skill between the guys that know C and the guys that don't. I work primarily in Java and certainly appreciate having spent a few years working with C before I did. On top of that I also did quite a bit of Perl work as well. I would say knowing as many languages as possible helps to give you different views on your work and applying different paradigms
C is not my language of choice, but even to this day, C is everywhere.
When I do some small code in Lua using LuaCurl, I use a C library. Lua itself is written in C.
When I do some Seaside Web application in Squeak Smalltalk, I use a VM generated in C (the Squeak VM is written in Smalltalk, and then it generates C code as a portable assembler).
So I would not start learning programming with C (see this thread for other choices), but as a programmer, knowing C is very handy even if it is not your language of choice.
In the same way that not every mechanic needs to know the inner workings of an engine to fix a car, not every programmer needs to learn C to produce code.
however, the ones who do, acquire a better understanding of the craft and ultimately achieve a higher level of success.
I share the same sentiments as the others in this thread, however, to answer the question that was asked:
The only programmers that have to know C are C programmers.
All knowledge is useful, so yes, you should envy their knowledge. You should also envy people who are AI nerds and know LISP, etc. The best mix would be a dynamic language, a functional language, SQL, a low level language and an object oriented language.
If you want some stranger to make some recommendations, I would go Python, OCaml, SQL, C and Java/C#. But, find your own path :-)
I think you have.
Languages have their own evolution. They developed within a very intriguing and fast evolution of computer systems. CPU power grew, features grew, Assembler got more complex... everything got more powerful.
Thing is: if you never saw the low level and "easy" beginnings, and you start with some high-level languages like C#, C++, or Java, you won't understand the elegance or backend perspective of these very powerful languages.
I think you don't need to learn LISP, because if differs a lot from common C-like languages. But some C is a must-know. It's for developers from developers, very near to machine code. Know what the machine does when you program it.
Memory allocation and pointers.
After getting to handle C and C++, even if only in school, you have a better understanding of what needs to happen in memory in order for you to throw objects and references around and you get a better appreciation of what, for instance, garbage collection implies.
Also, in school, starting with Pascal and then C allowed us to learn "programming" first, the old way, and then move to more advanced languages (OOP, etc) on top of that.
Have to know/learn C? Probably not, although it might make learning some concepts easier. Understanding something about memory allocation, structures and pointers of all kinds is worthwhile and C is a good language to use to gain that understanding. Plus, to be honest, "straight" C is really not complex. Tricky to get right, sure, but not in itself complex. I'd advise getting hold of a compiler that wasn't a C++ one too, that way can lie madness. (Fond memories of Quick C For Windows)
Have to know/learn C++? Definitely not.
I would use a metaphor: Knowing C for programmers is like knowing latin for (western languages) writers. It is not something you need, especially if you just write sports columns or cooking recipe books, but if you want to refine your craft it is something that I would consider nearly mandatory. But it will not be useful for daily work on ordinary software.
Or knowing mechanics for a car pilot, or how to build sails for a sailor. At some level of expertise, you need to know how the things you use are working internally.
If you learn C, try to master the pointer concept, and the way they map to the hardware. That's really the point of learning C. Do not spend time on the rest of the language.
Maybe it was just my particular educational experience, but I hated C/C++ in college and haven't touched it since. I'm thankful for learning about the concepts involved, like pointers and memory allocation, but trying to accomplish anything with the tools that were available to me was too cumbersome for me to want to bother. I hope your experience is better.
("slightly" tongue-in-check) You should learn it if, for no other reason, than it will make you love whatever language you're working in at that moment.
I don't call myself a C programmer, but I can write code in C. It has helped me a number of times in my career. I've spent a lot of time working with Visual Basic, and there are some things you just can't do with VB. It's been very handy to drop down to C to do things like windows hooks. It's made me the "hero" a time or two.
Closed. This question needs details or clarity. It is not currently accepting answers.
Want to improve this question? Add details and clarify the problem by editing this post.
Closed 2 years ago.
Improve this question
When Joel Spolsky and Jeff Atwood began the disagreement in their podcast over whether programmers should learn C, regardless of their industry and platform of delivery, it sparkled quite an explosive debate within the developer community that probably still rages amongst certain groups today. I have been reading a number of passages from a number of programmer bloggers with their take on the matter. The arguments from both sides certainly carry weight, both what I did not find is a perspective that is uniquely angled from the standpoint of developers focused on just the .NET Framework. Practically all of them were commenting on a general programmer standpoint.
What am I trying to get at? Recall Jeff Atwood's opinion that most of the time developers at such high levels would spend would be on learning the business/domain, on top of whatever is needed to learn the technologies to achieve those domain requirements. In my working experience that is a very accurate description of the work life of many. Now supposing that .NET developers can fork the time for "extra curricular" learning, should that be C?
For the record, I have learnt C back in school myself, and I can absolutely understand and appreciate what the proponents are reasoning for. But, when thinking things through, I personally feel .NET developers should not dive straight into C. Because, the thing I wish more developers would take some time to learn is - MSIL and CLR.
Maybe I am stuck with the an unusual bunch of colleagues, I don't know, but it seems to me many people do not keep a conscious awareness that their C# or VB code compiles in IL first before JIT comes in and makes it raw machine code. Most do not know IL, and have no interest in how exactly the CLR handles the code they write. Reading Jeffrey Richter's CLR via C# was quite a shocker for me in so many areas; glad I read it despite colleagues dismissing it as "too low level". I am no expert in IL but with knowledge of the basics, I found myself following his text easier as I was already familiar with the stack behaviour of IL. I find myself disassembling assemblies to have a look at how the IL turns out when I write certain code.
I learn the CLR and MSIL because I know that is the direct layer below me. The layer that allows me to carry out my own layer of work. C, is actually further down. Closer to our "reality" is the CLR and MSIL. That is why I would recommend others to have a go at those, because I do not see enough folks delving at that layer. Or, is your team already all conversant with MSIL?
Of course you should. The greatest way to become overly specialized and single-minded (and, correspondingly, have limited marketable skills) is to only work with a single type of language and eschew all others as "not related to your current task."
Every programmer should have some experience with a modern JIT'd OO language (C#/Java), a lower-level simpler language (C, FORTRAN, etc), a very high level interpreted language (Python, Ruby, etc), and a functional language (Scheme, Lisp, Haskell, etc). Even if you don't use all of them on a day-to-day basis, the broadening of your thought process that such knowledge grants is quite useful.
I already know C and that helped me during the 1.1 days where there are a lot of things that are not yet in the .NET base libraries and I have to P/Invoke something from the Platform SDK.
My take is that we should always allocate a time for learning something that we don't know yet. To answer your question, I don't think it is essential for you to learn C but if you have some time to spare, C is a good language to learn and is just as valid as any other language out there.
True, C is way below the chain. Knowing MSIL can help devs understand how to optimise their apps better. As for learning C or MSIL, why not both? :)
.NET developers should learn about the CLR. But they should also learn C. I don't see how anybody can really understand how the CLR works without some low-level understanding of what happens on the bare metal.
Spending time learning about higher-level concepts is certainly beneficial, but if you concentrate too much on the high-level at the expense of the low-level, you risk becoming one of those "architect" people who can draw boxes and lines on whiteboards but who are unable to write any actual code.
What you learn by learning C will be useful for the remainder of your career. What you learn about the CLR will become obsolete as Microsoft changes their platform.
My take is that learning some compiled language and assembly is a must. Without that, you will not get the versatility required to switch between languages and stacks.
To be more specific -- I think that any good/great programmer must know these things by direct experience:
What is the difference between a register and a variable?
What is DMA?
How is a pixel put on the screen (at low level)?
What are interrupts?
...
Knowing these things is the difference between working with a system you understand and a system that, for all you know, works by magic. :)
To address some comments
You end up having two different kinds of developers:
people that can do one thing in 10 ways in one or two languages
people that can do one thing in one or two ways in 10 different languages
I strongly think that the second group are the better developers overall.
I think of it like this:
Programmers should probably be actually working in the highest-level language appropriate. What's appropriate depends on your scenario. A device driver or embedded system is in a different class from a CRUD desktop app or web page.
You want your programmers to have as much practice as possible in the language in which they are working.
Since most programmers end up working on generic desktop and web apps, you want programming students to move into the higher level languages as soon as possible during school.
However, the higher-level languages obfuscate a few basic programming problems, like pointers. If we apply our principle of using what's appropriate to students as well, those higher level languages may not be appropriate for first year students. That throws out Java, .Net, Python, and many others.
So students should use C (or better yet: C++ since it's "higher-level" and covers most of the same concepts) for the first year or two of school to cover basic concepts, but quickly move up to a higher-level language to enable more difficult programs earlier.
To be sufficiently advanced in writing C#, you need to understand the concepts in C, even if you don't learn the language proper.
More generally though, if you're serious about any skill, you should know what goes on at least one abstraction level below your primary working level.
Coding in jQuery should be paired with an understanding of JavaScript
Designing circuits necessitates knowing physics
Any good basketball player will learn about muscles, bones, and nutrition
A violinist will learn about the interplay of rosin, friction, bow hairs, string, and wood dryness
I like to learn a new language every year. Not necessarily to master it, but to force my brain to think in different ways.
I feel learning C is a good language to learn about low level concepts without the pain of coding in assembly.
However I feel that learning lessons from languages like Haskell, python, and even arguably regex (not exactly a language, but you catch my drift?) is as important as the lessons to be gleaned from C.
So I say, learn about the CLR and MSIL on the job if thats your area, and in your spare time, try picking up a different language once every so often. If that happens to be C this year, good for you and enjoy playing with pointers ;)
I don't see any reason why they should. Languages like Java and C# were designed so that you needn't worry about the low-level details. That's the same like asking whether a WinForms developer should spend time learning the Win32 API because that's whats happening underneath.
While it doesn't hurt to learn it, you'd probably gain more from spending more time learning the languages and platforms you are familiar with, unless there's a good need to learn the low-level technical details.
It can't be a bad idea to learn MSIL, but in a way it's just another .NET language, but with nasty syntax. It is another layer down, though, and I think people should have at least some vague understanding of all the layers.
C, being somewhat like assembly language with nicer syntax, is a nice way to get an idea of what's happening on quite a low level (although some things are still hidden from you).
And from the other end, I think everyone should know a bit of something like Haskell or Lisp to get an idea of higher-level stuff (and see some of the ideas being introduced in C# 3 in a cleaner form)
If you consider yourself a programmer, I would say yes, learn C.
Many people who write code do not consider themselves programmers. I write .NET apps maybe 3 hours a day at work, but I don't label myself a "programmer." I do a lot of things that have nothing to do with programming.
If you spend your whole day programming or thinking about programming, and you are going to make your entire career revolve arround programming, then you better be sure you know your stuff. Learning C would probably help build a base of knowledge that would be helpful if you're going to go very deep in programming skills.
With everthing, there are trade-offs. The more languages you learn, and the more time you spend dedicated to technology, the less time you have for learning other skills. For example, would it be better to learn C, or read books on project management? It depends on your goals. You want to be the best programmer EVAR? Learn C. Spend hours and hours writing code and dedicating yourself to the craft. You ever want to manage somebody else instead of coding all day? Use the time you would put into programming and find ways to improve your soft skills.
Should .net developers be learning C? I would say "not necessarily," but we should always be dabbling in some language outside of our professional bailiwick because every language brings with it a new way of thinking about problems. During my professional career as a .net (and before that, VB 2-6) developer, I've written small projects in Pascal, LISP, C, C++, PHP, JavaScript, Ruby, and Python and am currently dabbling in Lua and Perl.
Other than C++, I don't list any of them on my resume because I'm not looking to be a professional in any of them. Instead, I bring back interesting ideas from each of them to use in my .net-based work.
C is interesting in that it really gets you close to the OS, but that's hardly the only level you need to know about to be a good programmer.
The CLR is a virtual machine so if that's all you learn, then you only know what's happening at a virtual level.
Learning C will teach you more about the physical machine as far as memory usage goes, which as you mention is what the CLR uses underneath. Learning how the CLR works isn't going to give you as much insight into, say, garbage collection, as learning C. With C, you really appreciate what's involved in memory management.
Learning CIL on the other hand, tells you a bit more about execution in .NET than you would by learning C. Still, how IL maps to machine language will still be a mystery for the most part so knowing some of the high-level opcodes, like the ones for casting types, isn't that helpful in terms of understanding what's really going on as they're opaque for the most part. Learning C and pointers, however, will enlighten you on some of those aspects.
Is the issue learning C or MSIL, or is it more fundamental? I'd say that in general, more developers could stand to learn more about how computers, physical or virtual, work. A person can get to be a fairly competent programmer by only understanding a language and API in a box. To take the profession to the next level, I feel that developers really need to understand the whole stack. Not necessarily in detail, but in sufficient generality to help solve problems.
A lot of these skills are being talked about here can be acquired by learning more about compilers and language design. You probably need to learn C to do this (whoops, sneaky), but compiler writing is a great context to learn C in. Steve Yegge talks about this on his blog, and I largely agree with him on this point. My compiler writing course in university was one of the most eye opening courses I've ever taken, and I really wish it had been a 200 level course, instead of a 400 level one.
I posted this on another thread but it applies here to:
I believe you need a good foundation, but devote most of your time to learning what you will be using.
Learn enough assembler to add two numbers together and display the result on a console. You'll have a much better understanding of what is actually going on with the computer and it will make sense as to why we use binary/Hex. (this can be done in a day and can be done with debug from cmd.exe).
Learn enough C to have to allocate some memory and use pointers. A simple linked list is sufficient. (this can be done in a day or two).
Spend more time learning a language that you are going to use. I would let your interests steer you into which language (C#, Java, Ruby, Python, etc.).
I was recently diagnosed with a cascading dissociative disorder that causes retrograde amnesia in addition to an existing case of possible anterograde amnesia. Many people have tried to remind me of how great a programmer I was before -- Right now I get the concepts and the idioms, but I want to teach myself whether I know or not. I think I can overcome the amnesia problems in part with it.
My question for you, stackoverflow, is this: I recently found Clojure and it... it feels good to use, even in just copying down the examples from whatever webpage I can find. My goals in learning a functional programming language are to create a simple webserver, an irc AI bot of some variety, and a couchdb-like database system, all of which lightweight and specifically for education. What flaws does Clojure have? Is there a better functional programming language to use right now for education /and/ application?
I think Clojure is a very nice language. If I should point to any defect it is that it's very new, and even though the language seems very mature and production ready, the tools and frameworks around it aren't. So if you are going to make, for instance, a web-app, don't expect to fire three commands and have a "Your first web app is running, now read this documentation to create your models"-page on your browser.
There aren't that many libraries written in Clojure yet either, but that's not a huge problem if you consider that you can use almost anything written in Java.
Haskell currently has a large following and a growing base of libraries and applications. It's also used for education and research. I find it a very nice language to use.
Haskell, Erlang and Clojure are all good choices. I would personally recommend Clojure, you might be able to do some interesting database stuff with the Software Transational Memory system that is part of Clojure.
You list CouchDB in your question, and it's written in Erlang, which is meant to be a pretty engrossing language once you get into it.
I have no personal experience with Clojure, but i really recommend F#. It's quite a powerful language in the style of OCaml. I really like it because it's debugging tools and IDE are second to none, and you can take advantage of practically every library on the (huge) .NET platform.
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
I attended a job fair yesterday and a developer asked me how I would rank my proficiency in C. I then realized that this is incredibly arbitrary and almost impossible to nail down, so my question is what knowledge makes you an expert in programming C?
Edit: or what would the breakdown be? what makes you good, decent, proficient, etc.
Edit again: I was looking more for like a list of skills or some other constructive measure by which to judge one's own proficiency in C, as that's hard to do.
List so far:
Experience in large projects
Mastery of Pointers (and memory management, I'd assume)
Mastery of a debugger (gdb, ...)
Mastery of a profiler (gprof,...)
Mastery of a memory profiler (valgrind, ...)
Knowledge of the fundamental standards
I think the trouble with this question is that the answer is kind of meaningless. I see people talking about experience, and that's good, and I see people talking about understanding the intricacies of the language, and that's good. However, if I were hiring someone to work on my C project, and I had a magic 8-ball that would give me an accurate answer to any one (and only one) question, I would never ask it, "Are they a C expert".
Why?
Just because someone is a C expert doesn't mean that they're a good software developer. Experience and language familiarity are good, but I think they are both trumped by that intangible, un-quantifiable property that makes someone a "good software developer". What I'm trying to say is, "What makes you a C programming expert?" is not a useful question, because there are more important questions. If someone is a Good Programmer, they will rise to the occasion.
As an example: You can be a C programming expert and be horrible on a team. You can be a C programming expert and refuse to use version control. You can be a C programming expert without knowing how to actually DO anything with C.
The "without" clauses in those sentences are equally important questions: What makes you a good team programmer? What's the best way to use SCM x or y? How do you approach programming a client/server game, or billing application, or web browser, or operating system, or compiler, in C? If a candidate told me "No, I am not a C expert", but gave me great answers to these other questions, I would hire them in a heartbeat over the guy who the magic 8-ball said was a C expert, but doesn't know how to check his code into subversion and hasn't learned a new language in 12 years.
When I interviewed with Google, the interviewer told me to think about it this way. On a scale of 1-10 for C proficiency, to say you're a "10" means you've written papers and/or books or been a speaker in a conference on programming in C. Based on this, very few people are 10s.
FWIW, I have been programming in C for 15 years. I consider myself very proficient. I'd perhaps give myself a solid 8 or 8.5.
To someone less skilled than you, you're an expert.
To someone more skilled than you, you're a newbie.
Experience is key, knowing the "rules" and syntax of the of the language is of course a must, but it is only a base. Learning the common pitfalls and idioms for doing things right is key.
Knowing what if any resources exist to get help from while your programing, and of course, knowing you're tool chain. I've known many C++ "experts" who had never used a debugger, or a memory tracker. If you ask me, being an expert in something is different from being proficient in something in you knowing all aspects of it.
Everyone is an expert at a job fair
You're an expert in c if you can answer all the questions tagged "c" on stackoverflow.com without blinking.
This doesn't directly answer your question (sorry), but it might help you decide how you classify yourself.
Instead of just "expert" and "clueless newbie" I prefer the three-level system of expertise used by the medieval guilds:
Apprentice
Still needs to RTFM.
Getting to grips with the tools and techniques of his craft.
Needs supervision.
Journeyman
Has Read The Effin' Manual.
Competent with all the standard tools and techniques of his craft.
Can work alone, and can supervise apprentices on routine jobs.
Master
Could have Written The Effin' Manual.
Is developing or adopting new tools and techniques.
Can oversee a major project that might never have been attempted before.
At a job fair? There are no experts: everybody's an expert. :)
Some may disagree but I think experience is key to being an expert in any language. I know plenty of people who've past the certification test but couldn't apply their knowledge to anything practical in the real world.
So I think overall being an expert is a product of having enough knowledge on a given subject (C) and then having applied it to enough real world scenarios to make the mistakes that we all do and learn from them.
The answers to this question do make for some interesting reading - it seems that we can't get good convergence on what defines an expert here. What hope is there going to be in a broader forum like a jobs fair? :-)
But to put my own 2 cents in...
I think there's two kinds of C expert.
There's the expert in the academic sense (as in "could write their own compiler", "has written papers").
There's the pragmatic expert. I would like to define this as "someone who can write elegant C code that anyone can understand".
I would take one of the latter over the former in a heartbeat. If you've got a chunk of code written by an expert that is so brittle that can only be read and understood by another expert then for all intents that code is unmaintainable.
It's all very nice that the author of this code remembers the intricacies of type conversions in the middle of expressions, but it's much better if the code has been written so that it's completely unambiguous.
Projects usually have enough technical challenge without adding the need for all team members to have memorized the C'99 standard.
You're an expert in C when you can write your own C compiler.
Interview questions like this are always tough. You want to blow your own horn a little, but not sound like a blowhard.
If you have done a lot with C (say, worked on open source projects in C), then I'd respond with that, but not just by pointing to the list of accomplishments on your resume, but by talking about one or more of them and what was particularly interesting or challenging about it (in regards to its use of C).
How about having read "Expert C Programming" by Peter van der Linden and remembering everything he covered?
Lacking a standard test there's really no way to decide what expert level is but here are a few of my litmus tests, everyone's list is different, I'm sure.
Without looking at documentation:
Know the precedence of the main operators so you don't have to litter your code with parens to avoid getting the wrong order of evaluation
Be able to write a prototype for a simple function pointer
Be comfortable with passing a pointer to a pointer
Understand block, function, module scope
There are more items like this.
On the other hand, I don't think you have to be able to understand or be able to write out Duff's device or figure out obfuscated C contests in your head to consider yourself an expert.
Even if I considered myself an expert (not sure I do) I probably would never claim it in a job interview.
Andrew
Mastery of pointers.
I would say that for any given language, experience is the key thing. It just takes time to learn a language and learn the APIs and 'idioms' that the language uses. Whether someone is an expert in anything or not is something that should be asked of ones peers. To paraphrase Jeff Foxworthy, "If you answer more questions than you ask, you might be an expert."
I think a fair answer would be understanding all of the intracies of ISO C.
The reality, as any comp.lang.c regular will tell you, is that almost nothing that people need to do can be done in pure ISO C, as you generally need to interact with your environment in a more well defined way. That's where POSIX comes about.
I would not blink at anyone who self-ranked themselves as an "Expert" who had a solid understanding of the language of C, a decent understanding of what ISO C promises, and a working understanding of the POSIX functions.
Being able to write papers/books doesn't necessarily make one an expert programmer. It takes plenty of hard work, practical experience and a good understanding of various C libraries.
Good luck!