Auto Code Commenting Software, Free Chairs

sien writes "When you think about it, code is usually fairly mundane and simple. Finally someone has come up with a parser and lexer that actually auto-comments code, allowing for vastly more rapid coding. This amazing new tool is called The Commentator and claims to analyse source code as it's being written and insert the necessary code comments. It's absolutely amazing. Also the problem of seating for eXtreme Programming has finally been solved."

LOL

[incubuz1980]

That is just so cool, asnd funny....

Re:LOL

[xstonedogx]

Woah, not so fast. We better consult TBS [tbs.com] on this one.

This is becoming tedious.

[reality-bytes]

Individually, these stories might be amusing but when they're covering the entire front page it becomes rather wearing.

Re:This is becoming tedious.

[Hamstij]

Hear Hear.

The fun of April fool's articles used to be picking the genuinely fake article amongst a whole heap of bizarre but true stories.

But this is just becoming boring.

Re:This is becoming tedious.

[Rei]

Perhaps it will be opposite this year? Picking the true article out of the heaps of really bad fakes?

Re:This is becoming tedious.

[yanko22]

Finally a good excuse for ./ to not have a single informative article the whole day. Woohooh

Re:This is becoming tedious.

[Anonymous Coward]

It could be worse: imagine what it would be if we had dupes about April fools !

Re:This is becoming tedious.

[Auckerman]

Every year slashdot does this, every year people complain. The joke is being fooled into complaining. The joke is people revisit the site like every normal day hoping for it to end and it doesn't. The editors are having a laugh at YOUR expense and you don't even know it.

Relax, go over to Fark, who's random joke page can actually be amusing and read about the Pope being note quite dead yet and come back to Slashdot tomorrow. If you actually miss anything real, you can read it then.

Re:This is becoming tedious.

[INetUser]

Well, if you notice that the comments count on the front page would appear to be at an all time low, so I guess /. has made itself irrelevant today.

Re:This is becoming tedious.

[null etc.]

I agree, but this one is actually pretty funny if you read it (and are a programmer).

short term memory loss

[commodoresloat]

I feel like the guy from memento. This is the third time I stopped what I was doing to look at slashdot, and it's the third time I forgot it was april fool's until halfway through the story. You think I'd get it by now.

Clearly, I need to recalibrate the delicate balance of caffeine and cannabis in my system...

Re:This is becoming tedious.

[jericho4.0]

I agree. Enough already. Some of us need to feel like we're wasting time more productivly.

Please shoot me now!

[nacturation]

Won't someone please think of the children!

Re:Please shoot me now!

[StalinsNotDead]

Please shoot me now!

Won't someone please think of the children!

Why? What are you planning on doing to the children?

Re:Please shoot me now!

[Atzanteol]

Perhaps he intends to *have* some.

chair..

[Kaisum]

That chair is not big enough for some men's asses.

1 Half Chair/Cheek ratio?

[seanvaandering]

Looking at it, it almost looked like it developed from that point alone ;)

Re:chair..

[ReverendLoki]

Sadly, that is the only chair big enough for some men's asses...

Re:chair..

[lpangelrob2]

They must not have designed that chair with XP in mind, because it looks like as soon as I sit on it the chair will lean over and break.

Come on guys! Test, build, *then* refactor...

Extreme XP seat?

[Anonymous Coward]

I sure hope this 2-person seat [cenqua.com] can handle a 500-pound couch potato, otherwise my partner will by flying.

I'm confused...

[AnonymousJackass]

That chair -- the "PairOn" -- is it for intimate co-workers, siamese twins or for people with big, fat asses?

Re:I'm confused...

[tmasssey]

One of the principles of Extreme Programming [extremeprogramming.org] is paired programming: two programmers sitting in front of a single computer. This is a chair to "facilitate" this, including being able to elevate the chair to "standing height" for the daily standing meetings, also a part of XP...

Re:I'm confused...

[zootm]

XP = eXtreme Programming, a means of development which preaches hands-on working and strict adherence to a few work processes. One of these is that all code should be written by two people at once.

Off-color Commentary

[Fox_1]

Beauty, this will replace the legions of QA people that used to do my code commenting

I know it's an APJ...

[jacen_sunstrider]

But I can't help but think how this could ruin some programmers. There are some programmers who live by documenting their work before they actually write it, as a guide to what they do. In fact, in my assembly language class, they say this is the best way to do it (not true IMHO, but oh well). besides, don't you just get a thrill out of documenting a finished routine, going to compile it, and realize you copy/pasted over a large chunk of code? or, more likely, forgetting an end tag and commenting out ha

Re:I know it's an APJ...

[JonTurner]

Just remember, Ninjas don't forget endtags like other mammals.

And if they DID wipe out half their storage, I would be like, "Dude! You'd better look out -- that Ninja just dropped the PENDING_ORDERS table in Production and I think he's going flip out and chop off somebody's head!"

good code

[theMerovingian]

Good code doesn't require comments - the variable names should tell you everything you need to know about the program.

Re:good code

[DarkMantle]

rriiiiigggghhhhtttt, and the tooth fairy created linux

Re:good code

[metlin]

I don't think that's a nick that Linus Torvalds would appreciate munnnchhhhhhhh.

Re:good code

[RichardX]

Good code doesn't require variable names - the memory addresses should tell you everything you need to know about the program.
And remember kids, GOTO is the one true flow manipulator. Accept no substitutes.

Re:good code

[lukewarmfusion]

That's right! Like this:

function MM_preloadImages() { //v3.0
var d=document; if(d.images){ if(!d.MM_p) d.MM_p=new Array();
var i,j=d.MM_p.length,a=MM_preloadImages.arguments; for(i=0; i<a.length; i++)
if (a[i].indexOf("#")!=0){ d.MM_p[j]=new Image; d.MM_p[j++].src=a[i];}}
}

(As an aside, this is a good example of optimizing code for production web delivery vs. clear, well-commented or self-documenting code for development.)

Re:good code

[lukewarmfusion]

In case it wasn't recognized (and judging by the troll mod, it wasn't) that is a Macromedia function. It's usually inserted by Dreamweaver (and maybe Flash or Contribute, both of which generate HTML pages).

Some of the things you claim I can't really verify - I don't know why they made the decisions they did or if there was any thought behind it at all. For instance, the indexOf part might be intended for support of a specific browser's implementation of Javascript.

This comes up in every discussion on comments...

[EvanED]

...and it's just not true.

Good variable names (class names, function names, etc.) go a long way: they tell you a lot about WHAT the program is doing. (I would argue that they can't always say everything too, but that's another matter.)

However, they don't tell you WHY you are doing what you are doing.

Also, remember there are other reasons for comments besides people reading your code. JavaDoc/Doxygen comments allow documentation to be produced right from the source. Comments such as /* FALLTHROUGH */ can tell sourch code analyzers such as Lint some useful information too. (Not to mention the programmer that looks at your code and has to think for a sec "did he mean to leave out the break there?")

this is why only one comment is necessary

[commodoresloat]

/* You're not supposed to understand this */

Re:This comes up in every discussion on comments..

[Tim C]

Don't blame the tool because the people using it are lazy and/or doing it wrong.

Re:good code

[bwalling]

Good code doesn't require comments - the variable names should tell you everything you need to know about the program.

You must be the guy that made the database at the last place I worked! The column names were complete sentences, including spaces and punctuation (it had never occurred to me that you could even do that). It was a real bitch to work with. Find the middle ground.

Re:good code

[kevinx]

I started to laugh, but then the laughter turned to concerned when I realized that this was modded insightful rather than funny.

Re:good code

[Moskie]

This is a joke, right? Insightful my butt.

Good variable names tell you what a variable represents. Is that the only thing you think comments can do? Perhaps for simple programs that's all you would need. But what about insight as to why a certain algorithm was chosen? Or documenting things like the implementation of business rules?

Admit it, you're lazy. I am too. I don't comment my code nearly as much as I should, but I recognize its usefulness and wish I did more commenting.

Re:good code

[Dracolytch]

I'm not sure if this is supposed to be FUNNY, a TROLL or someone just being STUPID.

It's April Fools, but I have met some people that actually believe this works. Those are also the same people that wrote the worst code I've seen in my life.

~D

Re:good code

[theMerovingian]

I'm not sure if this is supposed to be FUNNY, a TROLL or someone just being STUPID.

Welcome to /. on April 1, brother. Believe me, I was as surprised by the "Insightful" moderation as anyone!

OMG LOTF

[sfcat]

This is the first funny article of the day. I like the self-importance option. In truth, a tool like this is impossible to actually write, but it would be cool since I never comment my code. But I don't comment for job security and I hate my coworkers. Just kiddin'

Re:OMG LOTF

[Jerf]

In truth, a tool like this is impossible to actually write, but it would be cool since I never comment my code.

Well, if you punt on the relevance issue, you could probably come pretty close with an old-style Dissociated Press-style Markov chain generator and a sufficient corpus of comments to draw from. (The other parameters would affect what comments are used for the corpus.)

This could be sort of interesting. But then, pretty much anything fed into such a Markov generator is "sort of interesting".

I su

Extra Madden Option

[Shadow Wrought]

I wonder if it will read the code commentary aloud in John Madden's voice for extra money?

Re:Extra Madden Option

[Fulg]

I wonder if it will read the code commentary aloud in John Madden's voice for extra money?

Now THAT, my good sir, is genuinely funny. A definite "+5 Coffee up the nose", in lieu of my non-existent mod points.

Thank you for brightening up my boring /. fake story day!

Re:Extra Madden Option

[Shadow Wrought]

I was laughing as I wrote it thinking of all the "BAM's" and "BOOM's" that he would use to describe stuff;-)

Happy April 1st!

Re:Extra Madden Option

[schon]

sweat contest

Umm, exactly what do you think the two people are doing in that chair, that they would sweat so much? .. and why would *anyone* want to hold a contest to see who's the sweatiest?

Dead horse....

[matth1jd]

Beating a dead horse doesn't even come close to these stories now.... please stop. Please.

Re:Dead horse....

[rayde]

oh come on, it's april fools. if you want some real news, go to CNN's Sci/Tech page for today. we'll all welcome you back here tomorrow with duplicate stories on how much we all hate SCO.

That's an AFD product that would be of some use

[roberto0]

Funnier than Google Gulp, but not as funny as that Parrot gag from O'Leary a few years ago.

Quote from the Commentator

[Mad Marlin]

Think about it for a moment: You are that star footballer.

Ha! Fat chance.

auto comments?

[thundercatslair]

This is probably a joke, but comments are the most useful part of any code. I used to hate them finding it boring and useless. After a I wrote a few large projects I couldn't comment enough. Auto comments might not be as usefull rather then if you did them yourself

This should be a compiler option

[zigzag]

Along with the new auto-debug option.

Auto April Fools Submitting

[MadMorf]

Ok, obviously /. has implemented an Auto April Fools Submittal mechanism.

None of the /. staff are actually at work today, the AAFS is handling all of these submissions automagically.

Come on 12am...

Re:Auto April Fools Submitting

[Jellybob]

I'd never thought of it that way... I guess it would make sense though - they can arrange the yearly company party without anyone knowing they're gone!

It's genious really.

Comments...

[arose]

/* I don't know what I was thinking */
/* This shouldn't work... */
/* I'll just insert a backdoor here, no one will ever notice */
/* SCO code here, don't tell them */

Comments are for pussies

[govtcheez]

It was hard to write, it should be hard to read.

Nice example of poor commenting style

[Kelerain]

Very good humor! But on a more serious note, this is actually part of the problem with some comments. They duplicate the code. Comments like

i++; // increment i by one

really slow down the comprehesion level of most competent programers, because they have to filter out a lot of redundacy. Comment on purpose, on the more general function of things, etc. An automated program, could only really comment on the code that was there, and likely in an obvious way.

Not to detract from the marvelous humor of the 'article' but it was a good demonstration of the problem.

Re:Nice example of poor commenting style

[Peyna]

i++; // increment i by one

The problem isn't so much the redundant comments like this, it's that some people think that they are appropriate to tell us what the function does. Perhaps more importantly, "i" should have a name that tells you what it's function is, and the comment could tell you perhaps WHY we're incrementing it. If it is blatantly obvious so that even the least intelligent programmer can figure out the how, what and why of the function, no comment is needed.

So if you have:

i++; // increme

Re:Nice example of poor commenting style

[HFXPro]

what ID though? does application of this function have any side effects?

I can't wait!

[Stephen Samuel]

.... Until 12:00 Noon hits the International Date line.

It's 12:20 on the west coast. Only another 3 1/2 hours left.

(( On the bright side, though, I did convince a friend that I'd taken a job in Redmond as a shill for (OK: 'Spokesperson') for Microsoft. ))

got this from fortune

[mpaon]

Good programmers never comment their code; it was hard to write, it should be hard to understand.

I wish all the software I saw was as well commente

[Software]

I've seen plenty of OSS and proprietary software code that had comments that *looked* auto-generated. Stuff like:

/** Sets the Home to a new value */
public void setHome(String newHome)
{ ...
}

I've never understood why people do things like this. Why not do something useful: specify what's a valid or invalid value of newHome, say when it should or should not be called. Or just leave it blank if you can't find something useful to say.

Re:I wish all the software I saw was as well comme

[Vadim Grinshpun]

Because some places require you to write comments (even when they are really not necessary) and enforce it with a style checker.
One way to shut up the said checker is to generate comments of this sort. Sad, but true :)

Re:I wish all the software I saw was as well comme

[Oligonicella]

"One way to shut up the said checker is to generate comments of this sort."

Yeah, as opposed to providing a valid comment, despite it's being "not necessary".

Re:I wish all the software I saw was as well comme

[eyegone]

I suspect that you're trolling, but that is a Javadoc comment. The comment text, "Sets the Home to a new value," will be used as the method's description in the automatically generated HTML documentation.

Re:I wish all the software I saw was as well comme

[raygundan]

I agree. This happens all too often-- it's what you get when you mandate code comments but don't add time for it to your project schedule. Doing it right can frequently take longer than writing the code, and sadly, nobody will pay for the time it takes. So the programmers tag every "i=1;" with a "//set i equal to 1" and leave out anything even remotely subtle or useful.

Re:I wish all the software I saw was as well comme

[Oligonicella]

Lemme get this straight. You believe that the process of designing (or choosing) an algorithm doesn't provide you with the information you then simply type up as a comment?

Explain how that is possible, please.

In a working environ, a dipshit who fills their program with "//set i equal to 1" type shit deserves termination. It ain't for their own amusement, and it ain't amusing

Re:I wish all the software I saw was as well comme

[raygundan]

The information may be there, but organizing it and putting it down in a useful fashion is not something most developers ever do, let alone do quickly. It's rare that I see good info, particularly on side effects and non-parameter requirements (file formats, db requirements, external system constraints and expectations, etc...)

I agree with you that a good programmer should be able to do this instinctively after design and implementation of the code. Unfortunately, not every (or even most) programmers are

Re:I wish all the software I saw was as well comme

[InfiniteWisdom]

Possibly because companies have "standards" written by people that don't understand technology that dictate that every function must have a comment explaining its purpose. This can get truly atocious with OO languages where you often have trivial accessor functions like the one in this example.

Re:I wish all the software I saw was as well comme

[yomahz]

I've seen plenty of OSS and proprietary software code that had comments that *looked* auto-generated. Stuff like: /** Sets the Home to a new value */
public void setHome(String newHome)
{ ...
}

I've never understood why people do things like this. Why not do something useful: specify what's a valid or invalid value of newHome, say when it should or should not be called. Or just leave it blank if you can't find something useful to say.


Have you never used an IDE [eclipse.org]? They look auto-generated bec

The personality check answers all questions..

[Gopal.V]

if you have any more doubts - look at the configuration [cenqua.com] screen. Also peek at the commented code - looks familiar ?. (linux [linux.no], etc..)

"Commentator" - you can get it for free.

[AtariAmarok]

This is truly a revolutionary technology. There is even a mode whereby it will be nice and friendly and propagate itself through "Microsoft Internet Explorer" without you even asking: inserting comments into all the code on your computer for you! They can't do it all for free, so don't you worry at all about the ''' home loan and ''''viagra comments that appear in all of your code. It it is just part of their business model.

Enough of the damn April Fools jokes

[Laconian]

Slashdot has become utterly worthless today. I thirst for legitimate articles!

Re:Enough of the damn April Fools jokes

[nb caffeine]

Dude, thats every day. At least there isnt dupes today... (knock on wood)

Wow!

[ABaumann]

Man do I wish this was real. Maybe it could just take random quotes from bash.org and insert them as comments in my code.

Oooh! Brb... spending time at work writing code to insert random bash.org comments in commercial applications.

Utterly useless to me

[Cheirdal]

I can look at any piece of code in any project I've worked on and see what the code technically does. What I can't do is look at something and say "Ah, this criteria exists in the code because of business rule #275 for customer X." Almost all my comments are related to business rules and what I'm trying to accomplish. There is no way an autocomment tool could analyze the code for the business rules I use.

Re:Utterly useless to me

[Cheirdal]

Damn you April Fool's Day! ;)

Re:Utterly useless to me

[Jeremi]

Some of the most useful comments in my codebase look like this:

sLayout->addSpacing(10); // FogBugz #2768

This adds very little clutter to the code, and if anyone wants to know why that line is there, they can head over to the bug database and read all 50 paragraphs of description and debate regarding the problem that this line fixed, when it was added, how it works, etc.

Personality Controls

[Rightcoast]

I will run mine with verbosity turned way down and bitterness turned way up...Just like the real me!

This one is giving my PHB an idea :(

[mollog]

The PHB wants to order these so he has a place to sit while micromanaging.

Missing languages

[frisket]

Damn, still nothing for XSLT.

Auto Comment Slashdot

[superstick58]

I have adapted this technology to the Slashdot commenting system. Now, I will always have first post, always be the first to say In Soviet Russia...

I can already taste the Karma.

Let's see, it's April Fools...

[syphax]

... so it must be time to head over to ./ and BITCH about the lame April Fool's jokes!

Me, I love April 1 because I can ignore ./ for a full day. Or at least a few more hours than usual...

PaiRON?

[red_dragon]

PairOn... Pron... PairOn... Pron...

I don't know, but that chair seems designed exclusively for eXtreme Pr0n-watching with the SO.

Oh, wait; I'm on Slashdot... nevermind.

// TODO...

[duckpoopy]

//TODO: remove this comment

Cool Chair!

[lbmouse]

I want that chair! Finally a Herman Miller that will fit my fat ass.

best one

[jonathanduty]

This gets my vote as best one of the day.

Comments ... NO!

[jmartens]

Real programmers don't comment. If you did that someone might actually be able to figure out what you are doing. You need a code obfuscate function that takes simple functions and makes them complex and adds nonsensical comments, such as:

Before:
int i = 1;
while (i 1000) { ... A[i] ...;
i ++;
}

After: //Super Froopy node generator
int i=11; //Charge the Interociter
while (i 8003) { ... A[(i-3)/8] ...; //null node generator
i += 8;
}

But...

[RayDude]

If the comments are so simple that a machine parser can write them, how could they possibly be useful?

I thought the point of comments is to write the things that are not obvious. Obvious comments are coding spam where I work.

Raydude

Best April Fools I've seen :)

[Toby The Economist]

bitterness=9,profanity on

major lol :)

--
Toby

In the name of all that is geeky!

[Conspiracy_Of_Doves]

Enough! Make it stop! Make it stop!

First Test Run

[Nom du Keyboard]

10 PRINT "Hello World" 'what a novice
20 END 'boy I'm glad that's over

Groan

[Chanc_Gorkon]

Sheesh....Now the fscking ADS are even April Fools.

Who the hell is Dykstra?

[roman_mir]

The probably meant Dijkstra [utexas.edu]
but then again, they wrote the C0|\/||\/|3|\|7470r, who am I to question?

Re:ENOUGH!!!!

[jkmiecik]

Amen brother. It's just fucking annoying.

Re:ENOUGH!!!!

[jaysones]

I guess we'd better cancel Sense of Humor Day too, then.

Re:ENOUGH!!!!

[Jane Hackworth]

That...was "insightful"?

Would you F- off with the April Fool's whining already? We Get It! We've figured it out already: you are superior to the rest of us, who are obviously too easily amused.

To be bored with April Fool's Day is to be bored with life.

Re:Comments? Who needs comments?

[MyLongNickName]

It says "this guy shouldn't be allowed to touch himself, much less a computer".

Re:enough already, its past noon PST, stop

[Drooling Iguana]

Posting Slashdot comments via telegram?

Re:*THE* Authority for all april fools pranks

[DJStealth]

PLEASE MOD THIS (and other similar posts) DOWN, this has been in every single article today, and he has got mod points in each one.

See user's other posts at
http://slashdot.org/~Urgo [slashdot.org]