Fog Creek Software
Discussion Board

Too many bells and whistles

The WinAmp thread got me thinking about something that's been a problem in software for a long time but seems to be getting worse:  Feature Bloat.

This problem used to be only found in big programs, like word processors and database programs.  But now, even the simplest programs are now being loaded with more and more useless, pointless "features".  WinAmp is one example. 

Another is Nero.  It used to be a good CD burning program but now they've crammed way too much unnecessary crap into it. 

One of the former cool features was having Nero automatically launch when I inserted a blank disk.
But no more.  Now, when inserting a blank disk , your only choice is to lauch a "Wizard" that makes the CD burning process 10 times harder and more complicated.

CGI-bin Laden
Sunday, August 8, 2004

Just switch to normal mode. Click "Nero" at the bottom of the first Screen in the Wizard. Nero is one app. I have no complaints about. None.

Sunday, August 8, 2004

Unless customers are willing to rent the software, or the company can make money from other products once a software has all the features most of its customers are willing to pay for... I don't see how it can remain in business unless it issues new versions and get customers to buy them.

On the other hand, I'm not sure shareholders would agree on taking some of the profits of profitable products to pay for the maintenance of deadware. Also, how long can a piece of software remain usable once it's no longer under development because the company folded? I'm using Ecco as my PIM, and although it's still remarkably good considering development stopped in 1997, it's beginning to show its age.

Hence WinAmp 5 :-)

Sunday, August 8, 2004

I dunno, I like Winamp's new features, especially the way it handles playlists, and the aforementioned way you can mass edit ID3 tags.

Skinning I can do without, I've never been a big fan of skins, except on Trillian because the default skin is so ugly and difficult to use.
Sunday, August 8, 2004

Microsoft Works 2.0 for DOS is so simple and sweet!

Just Kidding

Li-fan Chen
Sunday, August 8, 2004

Think of it from a psychological point of view, too. Have you ever finished a software project you've spent a lot of time on? Do you really say "Okay, now I'm done"? Or do you think "Now that it's done, I can add that one setting I put off" or you get an email with a suggestion you really agree with...

Did someone once say "Software is never finished; it's only released"?


Sunday, August 8, 2004

In Eric S Raymond's "The Art of Unix Programming" one of his key points is: Do one thing and do it well.

The problem with this in the commercial realm is once it's done well enough, there is no longer any reason for users to buy or acquire new versions. That is a poor business model. Once it is good enough, then doing it better is not sufficient reason for users to make a new purchase.

So the alternative, moving contrary to ESR's philosophy, is to do another thing and glue it to the existing program in hopes of keeping the revenue stream flowing. This does indeed lead to bloat and extraneous features, but without it the income dries up and the company is forced to continue to create new things separately and absorb the expense of marketing the new stuff. Marketing costs for add-ons are significantly lower since your audience is already known.

Many other bloat "features" are not added functionality at all, but attempts to gain lock-in or lead-in to that company's other products or their advertisers. The basest form of this is collecting information on users - sometimes tracking cookies, sometimes spying on your system - and then passing it back for data mining and selling to advertisers and the like.

Monday, August 9, 2004

One exception to bloat for new versions is programs where the need for upgrades comes from external sources.

Not too many of those - tax software, maybe others that must deal with externally defined environments.

dot for this one
Monday, August 9, 2004

I dunno. I look at Winamp and I see:

1. A window with start/stop/pause etc.
2. An equalizer that pops out of it.
3. A playlist.
4. A place to manage all my mp3's and playlists.
5. A video window, if I decide to watch video instead of listen to mp3's. It is just a short hop from MP3 to MPG, after all, and it's convenient to be able to browse whatever videos I may have from the same interface.
6. A visualization window, so I can look at neat swirly effects while I listen to my mp3's.
7. There's also a web browser, mainly used to surf sites that will have streaming webcasts, or, more usefully, to display artist & song information, and with streaming webcasts, to participate in a community by rating or posting comments about the song.

Which of these is not "essential" to an mp3 player? Where do you draw the line? After 1, after 2, after 3, after 4, after 5, after 6 or after 7? Which of these qualifies as a "bells and whistles" feature that doesn't add value directly to the intended purpose of the software?
Monday, August 9, 2004

Who cares if the product has more features? I don't.

The only problem comes when they mess up the user interface.

Mr Jack
Monday, August 9, 2004

"The problem with this in the commercial realm is once it's done well enough, there is no longer any reason for users to buy or acquire new versions. That is a poor business model."

I'm not sure that it's a poor business model -- it works for the folks who make my fridge and washing machine.  They know that they'll sell one of these per household every 20 years or so and so they price and market accordingly. 

I may be the exception (especially in the Land of the Geek), but once I've got a piece of software that works, it takes a pretty substantial change in features/technology to get me to upgrade, even if it's free.  If I've got something that does what I need, why on earth would I change just to add something I don't?

Trollin, trollin, trollin, keep those dogies rollin'
Monday, August 9, 2004

So sell a subscription model like a certain Redmond based company that shall go un-named. "Upgrades whether you want them or not!"
Monday, August 9, 2004

There's a comment in _Future Shock_ about people renting their copier machines from Xerox ... and consequently:

People could simply stop renting (without losing their capital investment) if a much better machine came out next year

Xerox needed to keep innovating, to ensure that when the much better machine did come out next year then at least it would be a Xerox

Xerox were thus trying to obsolete their own technology, and assumed the risk and cost of technologies becoming obsolete (instead of the customers' assuming that risk, which they would if they were buy instead of rent).

Christopher Wells
Tuesday, August 10, 2004

And was this supposed to be good or bad for Xerox? In the long term it sounds good - they stay the market leader because they have no other choice.
Tuesday, August 10, 2004

> And was this supposed to be good or bad for Xerox?

The book didn't say if it was good or bad for Xerox. It was a footnote in a chapter that talked about how things were (sic) becoming more and more transient.

Christopher Wells
Wednesday, August 11, 2004

*  Recent Topics

*  Fog Creek Home