Geeky on NewsGator Online

Why Accurate Estimations Are Just Not Possible

Leon Bambrick — Wed, 18 Apr 2007 23:17:22 GMT

Estimating a new feature... it's like a little fish swimming toward you.

Here comes the little fish.

Look at the little fish!

The fish is tiny!

I can eat this fish! I'll swallow him in one bite!

Are you sure it's going to be so easy?

Of course it will be easy! This fish is tiny!

But you've been wrong before.

Shut up, voice in my head, this fish is tiny! I'm staring straight at it!

One bite. Munch! Gone! This will be a cinch! It's hardly a fish! Just a dot! A tiny dot! I'll eat the tiny dot!

Here comes the fish!

Wait a second. I'm getting a bad feeling about this.

It's too late. You're committed now.

Here's the fish from side on.

Oh. Damn.

Guest Article: Our Dirty Little Secret

Alex Papadimoulis — Fri, 23 Mar 2007 17:00:00 GMT

Hi, I'm Jeff Atwood from Coding Horror.

Alex offered me the opportunity to write an article for The Daily WTF while he was away on vacation a few months ago. Unfortunately, the scheduling never quite worked out, and the tone of the resulting article was different enough from The Daily WTF format that it never quite fit in. But now that the site has been relaunched, it’s a perfect time to revisit the question that has nagged at me ever since I wrote the original article. What are we all doing here?

I'm no stranger to bad code; after all, my blog is named after a sidebar illustration used to illustrate dangerous code in Steve McConnell's seminal programming book Code Complete.

But I'm not here to present yet another scary code sample by yet another developer who should have known better, but didn't.

Although I love Worse Than Failure like I love the world's ugliest dog, the concept has always made me a little uncomfortable. I've certainly written my share of WTFs through the years. And what exactly defines a WTF, anyway? One developer's clever hack is another developer's nails-on-the-chalkboard WTF.

Ultimately, I decided that Worse Than Failure makes me uncomfortable because it exposes the dirty little secret of our profession: we all write bad code.

The only difference between us and them, in fact, is that we know when we're writing bad code.

And that's a very thin line indeed.

There are millions of lines of WTFs written every year by developers who had no idea they were writing WTF material. The developers writing these WTFs are blissfully unaware of the carnage they're creating for their fellow developers.

And there's so much material to work with. In our profession, bad code is not remarkable, noteworthy, or interesting. Bad code is utterly mundane because it is, sadly, all too common. It's the status quo. If you want to see bad code, you don't have to come here to do so. Just look around you. This should make us uncomfortable. If anything, there should be a website where you could marvel at the truly unique code -- the stuff that doesn't suck.

Publicizing bad code is part of educating developers. Worse Than Failure has done that remarkably well. Pointing and laughing at all these terrible code samples is fun. But once you realize the scope of the problem, it stops being fun -- and starts being downright depressing. At some point you wish the guy before you had done something about this person and their WTF code before it became your responsibility.

Let's take a break from our regular programming to consider how we can prevent WTFs from happening in the first place. If you're reading this, you probably don't need our help. But your fellow developers do. Take less capable developers under your wing. Mentor them. Provide them with guidance. Invite them to a user group meeting. Email them blog links. Expose them to essential programming books like Code Complete and Don't Make Me Think. In extreme cases, maybe they legitimately picked the wrong profession and should be gently nudged into some other role.

Or you could just post their code on Worse Than Failure for us to laugh at.

But when you do that, you're really laughing at all of us.

Persistence Gets the Job Done

Alex Papadimoulis — Thu, 21 Sep 2006 16:25:13 GMT

Kristopher has the pleasure of working with Robert, The Infallible Programmer. Robert (his earthly name) is actually one of the world's two programmers who have never coded a single bug. The other bug-free programmer, though not infallible, is the guy who writes all of those "Hello World" examples.

This is not to say that the Code of Robert is bug-free. As a worldly programmer, Robert is constrained to worldly tools created by fallible programmers, and it is within these tools that the defects exist, and not His code. To Him, nothing is "real" in the sense of, "if we used a REAL database, this would work," or "if we used a REAL operating system, this wouldn't crash." But alas, He contends that no "real" tools exist on the marketplace yet: "everything out there is made by a bunch of idiots!"

But let's get back to Kristopher, who joined the company several months back. As an intern, Kristopher is in no position to question anything produced by Robert. Nor is anyone else: Robert is also the star-developer and His word is final. Besides, how could one possibly question the Code of Robert, anyway? It's infallible, after all.

Kristopher thought it would be an interesting exercise to show what exactly infallible code looks like. This example of the Code of Robert was found in the session termination subroutine of one of their web applications. Kristopher hasn't quite figured out what PHP, Postgres, Apache, or Linux bug this "trying the same database query 5,000 times until it works" code is working around, but I'll bet it's a pretty serious one ...

// Try 5000 times to make sure that this stupid thing 
// was deleted. This crap is such a toy.
while ( !$bDeleted && $iTimes < 5000 )
{
  
  $iTimes++;

  //does session exist?
  $sSQL = "SELECT * FROM sessions " .
          "WHERE s_user_id = " . $user_id . "";

  if ( !$qryObj->Exec( $sSQL ) ) 
    continue; //failed, try again
  
  if ( $qryObj->NumRows() == 0 )
  { //finally, deleted!
    $bDeleted = true;
    break;
  }
  
  //delete session
  $sSQL =  "DELETE FROM sessions " .
           "WHERE s_user_id = " . $user_id .  "";

  if ( !$qryObj->Exec( $sSQL ) ) 
    continue; //failed, try again
}

if ( !$bDeleted )
{
  trigger_error( "Delete failed.", E_USER_ERROR );
  die();
}

Plane Pr0n

lex — Tue, 19 Sep 2006 00:53:42 GMT

Courtesy of Brian, some plane pr0n taken from the flight deck of the USS Reagan, for them such as like it:

http://epiqs.smugmug.com/gallery/1101687/2/54850706

http://epiqs.smugmug.com/gallery/1101687/2/53160254

http://epiqs.smugmug.com/gallery/1101687/7/79151584

You know who you are.

No Tags

NewsGator Desktop Sync Beta (Windows RSS Platform / NewsGator Online Synchronization)

Nick Harris — Mon, 18 Sep 2006 22:06:38 GMT

Back in March at MIX06, Greg Reinacker and Waltor VonKoch demonstrated and blogged about their demo of synchronizing the Windows RSS Platform with NewsGator Online. Since then, IE 7 RC 1 was released with the finalized version of the Windows RSS Platform API and eventually the task of finishing and cleaning up our synchronization application and making it ready for the public fell to me... and now it's ready for you!

NewsGator Desktop Sync Beta is the evolution of the MIX06 demo and offers full synchronization between the Windows RSS Platform and NewsGator Online. Desktop Sync is a system tray application that keeps your feeds, folders and read states synchronized between NewsGator Online and the Windows RSS Platform. This means that any application that uses the Windows RSS Platform will be automatically synchronized with your NewsGator Online account!

You can download NewsGator Desktop Sync Beta here:
http://www.newsgator.com/download/desktopsync/ngdesktopsyncsetup.exe

Keep in mind that this is a pre-release, which means it's not done yet! So please don't download it unless you're comfortable using incomplete software that still contains bugs.

Requirements
Internet Explorer 7 RC1 or Windows Vista RC1
Microsoft .Net Framework Version 2.0

Any issues can be reported using the "Submit Support Request" option on the Help menu, or by posting to the Desktop Sync beta forum.

I'll post about some of the configuration settings later on, but for now go ahead and download the install and tell us what you think!

Troops test liquid armor

lex — Tue, 05 Sep 2006 00:31:32 GMT

It was only a matter of time, really:

Soldiers are to begin trialling a futuristic “liquid armour” that is worn like ordinary clothing but turns into a rigid shield as soon as it is hit by bullets or shrapnel.

The armour consists of material impregnated with liquid silica that has been modified using nanotechnology. It is designed as a flexible alternative to the current military armour, which consists of Kevlar material reinforced by heavyweight ceramic plates.

No Tags

Minerva - Single-player Half-Life2 mod

crankygeek — Fri, 23 Jun 2006 18:18:47 GMT

I’m a Half-Life addict. I pre-ordered Episode 1 and played through it in a few days. Fortunately, I spend a lot of time with Connor so I got to stretch the 4-5 hours of play time out a bit. All-in-all, I’m glad to have it, but it was too short. Of course, I’ll still pay another $20 next time…

So color me happy when I stumbled upon this blog post which introduced me to Minerva. It’s a third-party, episodic single-player mod for HL2. And it’s free. Can’t wait to load it up!

And in case you’re wondering why I’m reading blogs in German, I’m not. That blog is for the author of ituneswatcher / itunesspy which is the nifty little combo that allows me to post the music I’m listening to right here on my blog.

The Name Game: Naming Software Projects

jatwood — Tue, 09 May 2006 18:53:00 GMT

We recently had a spirited discussion here at Vertigo on internal project code names. We've used cities and superheroes before, but we needed a change. Scott gave us the following guidelines:

We like one word names.
We like them to be both pronouncable and easily spellable.
They have to be client friendly.
They should be globally unique (across Vertigo).
We like to have a set, from A-Z to choose from, in order.

The resulting discussion was a lot of fun. We immediately found Microsoft's list of project code names, and Apple's list of project code names, which are good starting points to get your creative juices flowing.

Here are some of the suggestions so far:

Types of Food
Video Game Titles (Atari 2600, Arcade, etc)
Brands of Beer
Roman Emperors
Cartoon characters / shows
Mythological names / gods
Cars
GUIDs (my personal favorite)
Gemstones
Types of Coffee drinks
States (and famous sites within states)
Counties
Streets in a city
Plants
Dinosaurs
Hitchcock films
Cheeses
Dog breeds
Colors
Trees
IRS Tax Forms
English monarchs
Famous People (eg, Sagan)
Wikipedia article names
Single letters (including unicode)
Radio alphabet
World War II era ships
IKEA product names
Birds
Beaches
Bridges
Web 2.0 names
World of Warcraft realm names
Types of Fasteners (nut, bolt, rivet, etc)
Ski resorts
National Parks
Mountain Peaks

Of course, no entry on naming would be complete without a reference to the classic Salon article, The Name Game:

In the end, however, attempting to quantify the benefits of a naming project may be just as small-minded as, well, attempting to quantify the benefits of a name. For the lucky client who truly clicks with his or her namer, the collateral benefits go far beyond nomenclature. There are new words to learn. Fun games to play. And, in the case of the Monkeys, unimpeachable warmth and love. "We got so much more than a name," says Robin Bahr of 98point6. "I mean, I got a name for my daughter. One of our senior executives identified strongly with 'Mescalanza.' No one calls him Jim anymore. His name is Mescalanza." Meanwhile, she says, "our senior manager for Internet development just fell in love with the name 'Jamcracker.' And so today, the Harvey meeting is known as the Jamcracker meeting. There are 300 people at this company who identify Jamcracker with Harvey."
Bahr claps her hands over her mouth. "Oh my God," she says. "I forgot. I shouldn't be mentioning these names to a reporter. Technically, we don't have ownership of those names. Jamcracker is still the Monkeys' property."
Bahr stops for a moment, as if listening to herself. Then she bursts out laughing. "Listen," she says. "I take it back. You write whatever you want to write. If someone out there wants to name their company Jamcracker, God bless them. And good luck to them."