So, people can't hardly write safe web apps in PHP without spraying XSS and auth bypasses and arbitrary shell executions and arbitrary SQL injections everywhere, and you also want to hand the attackers the ability to segfault your server or possibly even straight-up run arbitrary code?

Anyone smart enough to truly safely code a website in C is smart enough to learn a language to create that website which doesn't get on all fours and beg to be owned.

Yes, I mean that 100% seriously. If you're smart enough to safely write that C code, you won't. If you aren't smart enough to write that C code safely... and the first clue that you aren't is that you think that you are... you don't stand a chance. It's suicidal.

At my first web job circa 1996, our shopping cart product was mostly C, so I did web development in C for about two years.

I'm not sure that using C was the craziest thing about how we did things. Nobody really had an idea how to structure a web app, so it was more or less a dozen or so CGIs which read/wrote to flat files through custom-written dbm clone (woo, NoSQL in '96!). Lots of unsophisticated string munging.

Maybe if we'd had some better abstractions, it wouldn't have been as painful, but C code written to do a lot of ad hoc string cobbling is, in fact, a little crazy.

Despite this, the product pretty much worked and actually had a few hundred customers and a future.

(We even got looked at by Yahoo, but they bought some outfit named Viaweb instead.)

WHYYYYYYYYY

No, seriously, why? As far as I can tell, his entire argument is "web development in C is terrible and limiting, but not as quite as terrible and limiting as you might think." I guess that's good news if terrorists are forcing you to write websites in C, but I don't see even an attempt at explaining why you would choose to do this.

I've done a number of projects that were web-oriented in C and C++ over the years, and they are by no means my primary language. It wasn't that bad, especially if you take the time to learn how to do string manipulation properly (which isn't really that hard).

Come to think of it, even though we consider C to be "low level", that's really only a temporary problem that solves itself the longer you work with it and the more you build with it. It's a site different issue than the opposite issue of Ruby and PHP, which will never be able to have quite the raw performance of C.

It's a bit like the difference between fast-food and gourmet. Yes, fast-food American burgers is probably easier than fast-food... French ratatouille. When you're employees are barely passed childhood and not particularly trained in the arts, you'll bet your business on the burger (RE: the similar bubbles in the gastro-pub market vis technologicalist startup). But a practitioner at the top their game will do wonderful things with either.

I wrote a little backend section in C once to do a FFT on some data and it ended up being 7 times faster or so than the perl the rest of the backed was written in. It made sense there.

Developing a whole CRUD app in C however? Seems like it would feel like digging a swimming pool with a spoon.

A language like Python or Ruby gets you ease of coding and a moderate level of safety; a language like C gets you speed but at a high safety risk.

Haskell could get you often similar speed benefits, and there are even a couple of rather interesting frameworks for it (with even things like compile-time template checking), but the barrier for entry is unfortunately high.

Thus a language like Rust really comes into its own: speed kin to C with safety significantly exceeding that of Python and Ruby. Users of Python and Ruby may find the parts of its type system which must be specified explicitly arduous, but it does make it ever so much more likely that code that you write just works, correctly, which (as primarily a Python developer) I have found marvellous.

Rust isn't there yet for web development, but solid support for the HTTP layer is well in progress: https://github.com/chris-morgan/rust-http. When that's far enough along, then I'll get on to the framework I have in mind.

There are hardly any benefits in writing your web app in C ove r Java or Go. That 1% speed increase is nothing compared to the huge amount network wait these apps will be doing.

If you really consider it, what 95% of people write these days is glue between various services, and the parts that do matter, where you need the most performance, are already written in C. The reason why Redis, MongoDB and Postgres have good C bindings, because they themselves are written in C.

So when it comes down to it all you are really doing is string parsing and string transporting, thats really the last thing you want to leave to C. I haven't done straight-C code that has required me to do a lot of string manipulation, but I guarantee I'll probably leave a buffer overflow exploit open when trying to parse GET variables.

Or I can just do it in Java, and get all the performance and almost all the advantages of doing it in C.

I wrote a small (~300 line) C program to scrape an email and insert the part I wanted into a Postgres database. I used PCRE for the extraction and ECPG for the Postgres part. ECPG stands for embedded SQL in C for Postgres, and it looks like this:

    // Open the storage subsystem.
    void open_storage()
    {
      EXEC SQL BEGIN DECLARE SECTION;
      const char* database = DATABASE_NAME;
      const char* username = USERNAME;
      const char* schema = SCHEMA;
      EXEC SQL END DECLARE SECTION;

      EXEC SQL WHENEVER SQLERROR CALL quit();
      EXEC SQL CONNECT TO :database USER :username;
      EXEC SQL SET search_path to :schema;
    }

My takeaway was that it was a lot easier than I expected, but it was still kind of a minefield. ECPG isn't widely used or highly intelligent, so there are a lot of stumbling blocks despite the pretty high quality documentation. I would recommend hiding your ECPG behind some functions so you only have to run some of your files through the pre-processor. But it is a neat tool and I could see myself using it again.

I've always found it interesting that OkCupid is built with C++, although they started it before 2001. http://www.okcupid.com/about/technology

Why C? Performance. I wrote a C webserver on a PIC32MX795F512L (80 mhz/ 128kB RAM), and it was _screaming_ fast (for what it was running on). Easily handled ~100 clients making 2-3 requests/sec continuously to a JSON based API (I really don't know how far you could push it before it got slow), which is respectable for that chip ... but this was written to the bare metal with no OS (did use libraries for IP/TCP/HTTP/JSON, though). C isn't my first choice for web development either, but there's absolutely no way you could pull that off in a higher-level language.

You think? Cpoll[1] is actually on top of the techempower web framework benchmark [2]

onion to be honest looks more like a http server generation framework. similar to what rob fitzpatrick was talking about when he was advocating gos net/http. but then again what do i know i haven't tried any of them.

Personally, even though ruby makes my salary I like C. Always did. Treefrog looks the most like a modern web framework to me. [3]

[1] http://sourceforge.net/projects/cpollcppsp/

[2] http://www.techempower.com/benchmarks/

[3] http://www.treefrogframework.org/documents/tutorial

Might anyone looking for a speedup akin to C do better to look at Go instead?

I'm mainly a C dev and I would never ever use C to write a web app. Not because I don't think it could be done, not because I don't think it could be done well but simply because string manipulations in C are a pity. Containers which are not dumb arrays in C are a pity. SQL bindings are much more tedious than the higher-level languages counterparts (I wrote quite a bunch of C interfacing with SQLite, it works well but requires a lot of LoC compared to perl for instance).

Basically 90% of what you do in a typical web framework in your scripting language of choice becomes 10 times more tedious and error-prone in C (and probably 100 times faster but frankly, who cares?)

It's already painful when I need to write sysfs and debugfs interfaces (and the linux kernel has a pretty nice API to write in those).

The only reason I can imagine for using C in a web app is performances but then:

1/ I would try to profile my app and see where the bottleneck is. If it's really CPU-bound and C would really help I'd try to write a small C library for this particular piece of code and bind it in my scripting language.

2/ If I really must write the whole thing in C for some reason I'd try to push for a subset of C++ with at least std::string, std::map, constructors and destructor to ease ressource management and maybe exceptions for error handling. Done well you'll get similar performances and with a bit of luck you might actually come up with something remotely maintainable.

A lot of embedded systems have web configuration interfaces written in C. For example, your wireless router's configuration page is probably written in C.

It all depends on whether you've got a compelling reason to or not.

For us it was a simple decision, we have a mature stable product[1] that's written in C. So the path of least resistance is to write out own web framework (from scratch to avoid IP infringement problems). Sure it took time, a lot more time than grabbing something written by someone else, but that's not how it works everywhere (especially in corporate environments). Now we've got this library it makes adding HTTP client/server stuff to any of our C binaries relatively easy.

Likewise string processing in C isn't a problem once you've got a suitable library built up; ADT libraries (no I don't need to implement my own buggy hashtable again and again), regexp wrapper libraries, etc. They also help maintain consistency across platforms (our product ships on 6 different platforms including Windows).

If you're looking for instant gratification then C isn't the right language for you but a huge amount of the Internet is built on the back of things written in C. Are webservers (Apache, nginx) written in C similarly crazy? No.

Of course, it needs someone to write a solid C HTTP framework and then open source it so that it can be maintained/improved, now that would be useful. Sadly our employer has no plans to release ours, which is a shame but I have no control over that.

1. Coming up 20 years old now and >$1.5bn in revenues over those years, not too shabby.

Reasons not to use C for web dev: 1. Maintenance costs will go through the roof with C. C developers are not cheap, and most people who use C have are not web developers. 2. Hiring will be exceptionally painful. Again, most C devs are not web devs. Put another way, most web devs use other languages. 3. Some C developers are so good that very few people can understand their code (Linus talks about this, and I've seen it happen at a prior company)

By way of personal experience, in '99 we had an outstanding developer who was a rock star in C. When he left, I remember sitting in a meeting when an A+ developer (he really was good) was asked by the VP Eng, "Can you support his code". The response, "No". VP Eng: "Why is it not commented?". Response, "It's commented. In fact it's the best code I have ever seen. But I cannot support it, it's amazing, but its way beyond my skill level."

I have never seen a repeat of this experience in a 4th gen language. I'm sure it can/may happen, but have not yet.

Although we try not to talk too much about it, many users of Varnish Cache actually use it "because it is the easiest way to get a C layer in our web stack".

Others are proud creators developing web sites in it using both VCL (which is compiled into C) and C directly through VMODs or Inline-C in VCL. More:

* https://www.varnish-cache.org/docs/trunk/users-guide/vcl-inl...

* https://www.varnish-cache.org/docs/trunk/reference/vmod.html

* https://www.varnish-cache.org/vmods

* https://www.varnish-cache.org/trac/wiki/VCL

I once inherited a website written in C. I actually tried moving it to PHP but the server was hitting a rogue MySQL instance with a file socket and I couldn't find a PHP API to handle it.

There were two things I did that actually made it pretty painless to work with: I implemented a Dreamweaver template processor that you could hook up to output a string or execute a block on each template element. I also implemented a preprocessor that allowed injection of large HTML blocks:

    for (int i = 0; i < n; i ++) {
        [[[<tr><td>(((i))). {{{strings[i]}}}</td></tr>]]]
    }

Doug Crockford at the TC-39 panel last night talked about how one of the most exciting things about JavaScript is that its a wildly successful general-purpose programming language that "doesn't have to look like C to be successful."

His point was that we used to live in a software ecosystem where the conventional wisdom was that if a programming language didn't look like C, it wasn't going to take off. JavaScript (and even more so variants like CoffeeScript) are showing that we've moved on from that sentiment.

In some ways, this article feels like a retreat to that older assumption. As if we somehow are expected to be searching for ways to work C into the modern web app landscape. Perhaps this point may somewhat orthogonal to the topic of the article and the related discussion, but I can't help but ask what problems are we solving by going back to C for web development? Is the assumed performance gain going to outweigh all the other benefits of using a higher level abstraction?

None of the framework examples listed in the article really seem elegant at all when compared to higher level counterparts (Rails, Django, Spring, et al). Of course, I'm saying this without having attempted to build an application with any of them.

undefined

I don't want to start a language war here but as a phd student working on distributed systems, choice for "C" as a language for developing high throughput web applications has some rationale. For instance, when you read most of the articles in HS.com or C10K at least, web development at large-scale requires every performance bit out of machine. Indeed, some -caugh-google-caugh- still use C++ at backend (even not in the frontend) or we see twitter switched to JVM from RoR for performance.

For regular earthlings, RoR, Py, PHP and alike are still more feasible than C for "getting the job done".

But the article's point of view is a bit shallow in that sense, since it's looking a way of replacing conventional frameworks (RoR etc.) for C. If you gonna develop a high-perf application in C, you probably use more wide variety of specialized libraries for focused tasks(like json processing, I18N, handling SSL, developing nginx/HAproxy modules,extensions for load balancing, etc.) rather than using a single "C framework".

undefined

The Fossil DVCS has a built-in web interface for reviewing commits, along with a bug tracker and a wiki. It's all written in C, albeit with a couple of custom macro preprocessors.

I once wrote a toy web page in C; Lex made for nice HTML templating and with pseudOO it actually felt pretty modern, but I still wasted a whole lot of time reinventing various wheels.

Well, C++ already has a couple of web frameworks : http://cppcms.com/wikipp/en/page/main

http://www.webtoolkit.eu/wt

I'm not sure that's a good idea though. It's hard to fail safely in a language that can fail silently. And failure modes include, not just blatant hacks into the underlying system, but surreptitious insertion that can propagate to all your visitors.

Besides the fact that it's pretty hard to find a decent C developer these days who's also interested in web development, you have a climate that's not exactly conducive to development in it. If it breaks or needs upgrading, will you do the work yourself? If it's that small, then it's probably not something you'd worry about too much, but then it's only on your shoulders.

Also, OKCupid is written in C++.

If you do want do do it in c, I think php pretty much got it right -- write an extension in c, and use something else for the rest. Se eg: http://phalconphp.com/en/

Then there's of course G-Wan -- source of a lot of speculation, and frequent border-irrational (counter)claims by it's author -- but undoubtedly a good contender for easy-to use server with c-language "scripting":

http://gwan.com/

A newer server in the same space is nxweb, which seems made for doing the kind of work the article talks about - simple, high performance web-services in c:

https://bitbucket.org/yarosla/nxweb/wiki/Home

For other languages, have a look at hello-world webservers at rosetta:

http://rosettacode.org/wiki/Hello_world/Web_server#C

Other related projects:

mongoose - small embeddable, cross platform web server w/lua scripting: http://cesanta.com/

lighttz - webserver in c using libev: http://arekzb.wordpress.com/2008/04/27/lighttz-a-simple-and-...

Still, if it were me, and I needed "c-speed", I'd probably go with cgi (or fastcgi). Forking on every request might not be the absolute fastest, but with the binary cached in ram, it's still pretty fast on Linux. Or maybe just use mongrel2 and a c handler:

https://github.com/derdewey/mongrel2_c_handler/tree/master/l...

Why? If you want the speed of compiled code you can come very close with .Net or JVM based solutions.

Writing web services in C can of course be done; I even believe that there are legitimate designs were it makes sense. Namely, if the back end is anyhow custom written and all you need is a smallish server to talk to the client. ( For example games, were the server simulates a world.) But the problem I have with the article is, that in this case frameworks don't help. Web applications were C makes sense, require a very different coding style from the one the article deals with. So it makes no sense to fetch just some values from a DB and give it to a server using C. C only makes sense if one has to do a large part of the heavy lifting oneself.

This is a nice summary, and the links are handy reference for the subject.

Security concerns aside, I've occasionally used small C routines in web-app CGI calls, and often wondered if others have gone further with C and web development.

You can have the best of both worlds by writing low-level modules in C for use within a higher level framework.

For example, I've been having a great time writing this music player backend in C[1], but as soon as it's solid I'm going to write a node.js addon module and then use that as the backend for my web-based audio player[2].

[1]: https://github.com/superjoe30/libgroove

[2]: https://github.com/superjoe30/groovebasin

Sure, cool stuff and probably great fun doing it.

However what matters in web development is not performance, nor that it is bug free and not even security. All of those things can fixed one way or the other on the server with more hacks.

The one thing that matters is speed of development, that is why PHP rules so hard.

"Let's argue about whether Haskell or Clojure is better while somebody else ships products using PHP and duct tape."

https://twitter.com/agentdero/status/174965036928868352

You missed to mention Duda I/O:

   http://duda.io

Even is do in C, i tries to present a friendly C API as you can see in http://duda.io/api, if you think into a fast big data server front-end, Duda I/O fits pretty well. A few slides:

    http://duda.io/slides/

The main problem, I think, is that web development usually requires lots of string processing, and the C standard library makes this complicated and opaque. Maybe Cello helps with that?

No, not crazy. Just really tedious. I've been there and even webdev in C++ was magnitudes more comfortable than webdev in C.

But webdev is a solved and trivial problem. More so it is an I/O-bound problem. So you can use the slowest language in the world as long as it boosts your productivity or makes your life super easy. (Sure, there are edge cases but 99.9% of webdevs never will be in the situation of having thousands of requests a second.)

We wrote Bloglines, the first and greatest RSS reader web app, in C. And by "we" I mean Mark Fletcher did, and the people he hired like me, continued it.

It wasn't bad. You just get used to keeping track of memory, arrays lengths and pointers like you do anything else. While I do prefer Python these days, I always enjoyed the looks of horror Java and PHP programmers greeted me with when I mentioned our language of choice for Bloglines.

If you want to make an Arduino serve websites, you'll have to write embedded-C commands that write HTML to browsers and can interpret POST commands.

undefined

I would like to add that apart from being professionally chosen or not, this sort of opens up a way for learning web dev concepts for those who learned programming with c as the first language. They just get to see internals of all libraries quickly rather than having lots of frameworks with no easy access to source to see what is going on underneath. Good idea.

It seems like if you want to get the performance benefits of C with web development, that's something you can do, but it does seem downright crazy to try to do it all with C. Use something safe and abstract as glue and for talking to the client, and use C for any heavy lifting, if the benefit can justify the extra work.

i been working with ntop, its a web interface written in C, the HTML is embedded within the C code. As a rails/php developer this is kind of crazy :P Heres the source https://svn.ntop.org/svn/ntop/trunk/ntop/httpd.c

The reasons for this [that no one uses C] aren’t terribly difficult to discern. Let’s face it: compared with higher-level languages, C is hard. Like, objectively hard

So.. instead of leading with one of the many, many great reasons not to use C, he pretty much calls everyone too stupid to make it work.

The worst abomination I've worked with was a web app written in PL/SQL. It was lovely to maintain.

"And when was the last time you wrote a function that returned 1 or 0? If you’re a Rubyist or a Pythonista or a server-side JavaScript devotee, you may never have done so."

That's just returning True or False (bool). Lot's of people return that from functions in all languages.

I am using the atozed intraweb framework for web dev: http://www.atozed.com/intraweb/index.en.aspx Comfortable C++ framework with session management and all the other stuff you need

When you get 90x the performance using C rather than PHP, the only thing stopping its use is human resources. To put it into perspective, you would need 90 PHP servers to do the same work in the same time as a single C server. Any more debate?

In the early days when Perl was the CGI de jure the company I worked for wrote their cgi in C and ran on Zeus webservers. This was for performance on one of the most highly traffic'd sites at the time.

Cgit is written by C and it's quite easy to develop with a fairly clean code base:

http://git.zx2c4.com/cgit/about/

Sorry using C is a security nightmare. You need to use a higher language like C++ that can encapsulate and abstract your web-framework enough.

Or use a C framework that has secure string handling and others.

Slightly offtopic, but I love Node.Native[1].

It provides an API similar to Node.js using C++11 lambdas.

1. https://github.com/d5/node.native

Also worth mentioning, a Websocket server library in C. https://github.com/m8rge/cwebsocket

Sounds like a fun side project. But I would not risk doing such thing for money - support would be a nightmare. And by the way, wasn't that what Go was developed for..?

He forgot the use case for C web development. Small embedded systems. Not everything is CRUD.

libfcgi (C) + postgres worked like a charm for me. There are also things like libmicrohttpd, libevent/libev, ... that are also very useful if you want to play even more. It's not that hard if you know C.

why use C? if you want to go down that path why not use C++ frameworks, like http://www.webtoolkit.eu/wt ?

Weather Underground is implemented in C.

Yep. Just bat shit crazy.

Not crazy, just stupid.

Yes.

I don't think the author has offered a great argument in this article. Building a new website in C sounds like a fun learning experience, but it's not a practical choice for the typical startup CRUD app.

When you're a startup, your biggest problem 99.999% of the time isn't running out of memory or not having a fast enough app, but getting an MVP up and launched and finding enough paying customers so that you don't run out of money. Except in specialized cases where your app is very CPU intensive, C provides a major headache without a major immediate advantage.

But C is a great choice for replacing parts of your app if you start growing and when you start figuring out exactly what services need to be faster and use less memory.

I own a web dev company. We do everything in C. For 10 years we've been doing so. The advantage is, we can do anything and everything. We do it faster, too. And we never have to wait for the language or some library to catch up to the new stuff. There is nothing we can't do and speed or platform is never a limitation. In fact, there's always this underlying feeling that coding in anything but C is crazy.

undefined