Software Code Quality Of Apache Analyzed

fruey writes "Following Reasoning's February analysis of the Linux TCP/IP stack (putting it ahead of many commercial implementations for it's low error density), they recently pitted Apache 2.1 source code against commercial web server offerings, although they don't say which. Apparently, Apache is close, but no cigar..."

Comments

So if they found them...

[Marx_Mrvelous]

Why don't they fix them? It seems almost paradoxical, if you find .53 errors per thousands lines of code and fix them, then you'll have 0 errors. But since we can only fix errors we can detect, we only detect errors we can fix. Ok, it's too early on a Monday morning...

Re:So if they found them...

[aborchers]

It's an argument to a function. It cannot be modified by another thread/process.

Thanks for the reality slap. Years of LISP and Java have made me weak and flabby. :-)

Re:So if they found them...

[Alsee]

I don't think using a development branch is really a good choice at all. Dev branches are just that, development, not intended for normal, every day use (except by the very brave).

Some people love the thrill of skydiving and opening their parachute 5 seconds before they hit the ground. Some people defy death by wrestling crocidiles bare handed. Others get a rush pushing 200 MPH going into a turn in a formula one race car.

Me, I get my adrenaline pumping by running code on the development branch.

-

It's not fair!

[jpmahala]

Just because Open-Source coders can't spell when they insert comments doesn't mean that they can't write good code!

Re:It's not fair!

[MrPerfekt]

Unless they can't spell other things like...

inklude
dephine
retern
brake... etc.

Re:It's not fair!

[Jucius Maximus]

"(putting it ahead of many commercial implementations for it's low error density)"

This line gave me a good chuckle. I expect that most people did not even notice the grammatical error in a sentence talking about low error densities.

Note: The rules for its/it's are not covered in Bob's Quick Guide To The Apostrophe, You Idiots since the Guide covers nouns and 'it' is a pronoun.

their own code?

[Jearil]

Why does it seem a bit odd to be testing software quality with other software? I wonder if they ran their own software through its own program, but then that gets kinda weird when a program starts noticing errors about itself... maybe it'd get depressed and start ranting at the creator on how they should have taken better care of it... ok, I need more sleep

Re:Code defects appear to be a small part of the e

[demaria]

Hypothesis: Taking down IIS, Windows or Microsoft is more fun/cool.

Re:Defect?

[Anonymous Coward]

so what are the calling a defect?

I guess would be quite a good example.

Re:Apache 2.1...?

[Bu Na Dan]

the error density in the announcement of reasoning.com is pretty high ... testing a non released software against an unknown commercial software ... sounds like an ancient tale. where are the people who accept this kind of crap?

BSD codestyle...

[BigBadDude]

The defect density of the Apache code inspected was 0.53 per thousand lines of source code...


We can bring this number down to 0.2 by avoiding the BSD style guidlines. No kiddings, have you seen the density of MFC code?

BSD code:

char*
foo(int bar, double baz)
{

/* do something */
return bar + random();

}



MS code:

char* Foo(int nBar, double dBaz) { return bar + random() + m_ExtraWindowsBugModifier(); }

Apostrophe abuse

[worst_name_ever]

putting it ahead of many commercial implementations for it's low error density

...but behind the Slashdot editors in terms of number of abuses of the word "its" per story.

Re:FACT: 3 is a larger number than 2

[rootofevil]

Turing says no.

If Apache is so poor in quality...

[tsetem]

...then why is it their webserver? :)

Of course it is Apache 1.3.23...

Recursion

[sterno]

They didn't do that because if they did that, then they'd find bugs in their bug finder, so they'd have to run the bug finder on the bug finder to find bugs there, but then they'd have to run the bug finder on the bug finder on the...

Re:Recursion

[fgb]

That reminds me of an old (early 1980's) product named BILF (Basic Infinite Loop Finder). It was supposed to be run against BASIC source code and it would find all infinite loops in the code, or so the vendor claimed.
A magazine reviewed the product. In their review they included a formal mathematical proof that such a program could never work. The vendor responded to the proof by saying that they would fix that problem in the next release!

In other news... I have begun testing

[teamhasnoi]

Apache 4.2 Alpha, a release that is yet to be even a twinkle in it's Daddies' eyes. I have found a whole bunch of errors, bad comments, a few scribbles on napkins, some old Populous save games, and a letter to 'Mom' asking for money.

I compared this to my 'other' server, for now unnammed.

My 'other' server brought me coffee, 2 pieces toast, 2 eggs OVER EASY, 4 strips of bacon, *and* Smucker's Grape Jelly with nary a mistep, or hesitation. This other server smiled, asked how my wife was, and brought me a new fork when I dropped my first one.

Congratulations, Gloria! You win the 'great server' award!

This article isn't worth the 2 dollar tip.

Here's an idea

[Daath]

Why doesn't Reasoning fill the niche, and code a completely error free web server? They know other peoples mistakes, so they should know how to code an error free one.
Well, seriously, I wouldn't put much in their obvious estimation.

Re:Bad Statistics...

[Lxy]

My general rule is that if someone is quoting statictics to you, they are lying. At least on average. :)

39% of Slashdot readers already know that.

Re:Wait a second

[willamowius]

They probably compared it to IIS 7.4 to make it a fair comparison. ;-)

every program.

[leuk_he]

The # flaws per leads to:

-Every program can be at least one line shorter.
-Every program has a least x bug per xxx lines.

Conclusion:

The ideal program has no lines and no bugs.

and to prevent any insightful moddings of this post:

Yes, the design is more important than the quality of the software, ask MS about this.

Re:every program.

[Asprin]

Except I heard it as:

Since every program contains at least one bug,
and further, every program can be reduced by one line.
Therefore, by induction, every program can be reduced to one line which doesn't work.

The proof is left as an exercise for the reader.

sco!

[Anonymous Coward]

The lower defect rate in Linux TCP/IP can only be explained by a large chunk of more mature, commercial, stable SCO UNIX code.

Thank you, Captain Obvious

[Sxooter]

Well this certainly falls under the "duh" category. Freshly written code tends to have fewer bugs than older, well reviewed, well tested code.

Wow, next we'll learn how you shouldn't buy any Ford, GM, or Chrysler product in the first year of production.

Re:what is a "software error"?

[beta21]

These acronyms sometimes get me IMNSHO?

I Am Not A Single Horny Octupus?