The Recursive ISV

If you're new here, you may want to subscribe to my RSS feed. Thanks for visiting!

As I’m still polishing away at MixAction and not ready for release, still, the posts I’ve been making here recently have been more general in respect of mISV issues.  This post continues that trend, I’ll be talking more about progress on MixAction later this week.

Recently on The Business of Software forums a statement was posted “Idea clearance“.  A few suggestions were put forward for software ideas should somebody looking for a product care to pick them up.  I’m not going to say here these ideas were bad, that would be untrue and unfair on the person posting.  But I am going to say that if you need these ideas for creating a new product you really ought to be considering not starting up - period.

Sounds pretty harsh.  But honestly, the world does not need yet another media player.  As a basic device aimed at consumers, serving no more purpose than iTunes, WMP or the horrifically concept challenged WinAmp - a product, it could be argued that has gone from mediocre to sensationally wretched. 

Now, I’m only using the media player as an example here, other suggestions were made, but to my mind this is the one that sticks out best in terms of want to be mISV’s looking for a product and choosing a sure fire loser.   Funny coming from a guy writing a special purpose “media player”, but the catch is it’s special purpose.  It’s not just a “media player” and does not directly even try to compete with the “big three” detailed below.

The poster suggested some ideas for such a player that included:

• Isn’t bloated
• Fast and resource friendly
• Catalogs my collection efficiently
• Flexible searching
• Can play everything under the sun, including streams
• Find duplicates

I’m going to deal with each of these in this post.  To say they are entirely subjective is an understatement.  In fact trying this is a recipe for mISV disaster, and I’m going to give reasons for why on that too.

First, before anything else, it’s got to be understood that competing with applications that come pre-installed on an OS is starting off on a foot that only a three legged person possess.  Now if you actually are Jake the Peg maybe you do have a chance.  But let’s face it, how many of us really do have that extra leg?

Sure, you might pick up some connoisseurs who may indeed love it - but can you live off of it? 

I mean that’s the goal isn’t it? 

The “V” in ISV stands for “vendor” and a vendor vends, charges, makes money out of a product or service.  In fact the only way you’re going to make money out if it, real “I live on the proceeds money” is if you can sell it to a wide enough group of people. 

Frankly I would argue you have Buckley’s hope of really making it with a product like this - a mere “media” player fitting all or any of the criteria above.  Ten years ago, sure, you would have been “there” for the birth of Win-Amp”, or the the development and death of Audion on the Mac (the “almost” iTunes predecessor and that link, Audion above, is an excellent read too, don’t miss it). 

Alright - who the hell am I to make this proclamation after all?

Fair question.  The answer is a person who’s watched developer after developer bash themselves onto the rocks of economic extinction because they didn’t see the lighthouse.  I moderate three “shareware” groups on Usenet and have seen developers come and go.  In far to many instances they were trying to market something like this “media player” suggestion. 

Keep in mind it’s a snap to clone too.  The 30 Day challenge, blogged about and something that I participated in here during June would be the kind of time period a newbie developer could produce such a product.  But then along comes somebody with better coding skills and does it in under a week.  Goodbye USP and if that person is a better marketer to boot you’re screwed, even if you really did have the “extra leg” we talked about above.

Don’t think so?  Try it.  Post the challenge on a blog and tell us about what you’re doing.  Then watch the “attack of the clones” - even if you can sell enough to pay for server costs - and I firmly believe you won’t.

With the opening gambit above we’ll look at the “features” next.

1. Isn’t bloated

   Lets see now.  Define “bloated”.  Is that anything over 50k, 100k, 300k, 1 meg, 3 meg?  If you’re using .Net you’re pretty much out of the game by those measurements already.  VB?  Ditto.  Delphi?  Ditto, though you could use Delphi with KOHL (good luck with that one).  Java, nope.  C++?  Sure, it could be done.  But how long to code it?  Assembly?  How long have you got again?

   Basically “bloated” is entirely relative in the context of size of the executable.  “But I’m talking about feature bloat” I hear you saying?

   Fine.  To many features spoil the broth, or some such twisted analogy.  But out of the six gotta haves for this “idea” we have three elements of bloatware features. 

   A) Catalogs my collection efficiently
   B) Flexible searching
   C) Can play everything under the sun, including streams

   If you know your Audio and indeed programming you know automatically that cataloging with flexible searching is going to require some kind of local database, whether it be something tiny like SQLite, something proprietary of your own invention or another third party solution.  XML is a possibility, but it’s a cow to parse on the desktop and not as fast as a dedicated local dataset with indexes.  Good luck on caching it too if you need to read just portions in.

   So we have “bloat” there. 

   So far we haven’t even gotten to the stuff like the UI, and I’ve not looked at item C yet - Audio formats including streams.  Coding your own code to access “everything” under the sun is no mean feat.  Indeed some of those file formats prohibit you doing so unless you use their libraries, which means DLL’s (or your platforms equivalent), plus the whole DRM debacle and coding for that (which you are going to have to pay big dollars to do legally)  You have to code for DRM to “make it” with this app because if you don’t you’re not in the game with the big three, iTunes, WMP and WinAmp.

2. Flexible searching

   How?  As above a database is going to be the most efficient method in terms of time investment and coding ease, reinventing the wheel is, well, pretty lame at this point. 

   But what kind of searching?

   Searching on every element of an MP3 file?  Which version - IDv1, IDv2 or IDV3 (and their branches)?  Cool for cats, what about Ogg?  Whole new tags to parse.  Then there’s APE, Flak and a bunch more. 

   The bloat is creeping up now… 

3. Can play everything under the sun, including streams

   Whoo-hoo!  Now we’re really in bloat-ville! 

   Taken a look lately at what “everything under the sun” in terms of Audio file formats actually means (note - this post hasn’t even mentioned video formats which of course would be included in a “media player”).  They are a massive list.  Some are proprietary and can’t be licensed at all, some are specialist, like the Acid Pro loop format that is an extended version of WAVE format. 

   How many bits?  8, 16, 32 or 64?

   Then you’ve got a bit rate.  A nice long list there.

   Streaming?  What.  Icy?  There’s quite a few of those out there too.  More “bloat”.

4. Find Duplicates.

   We’ve not even included the code to play the Audio files yet, let alone video, nor the niceties Joe and Jill Sixpack in their endearing diet of clickware addiction demand like graphic EQ’s (how’s your logarithm knowledge?), compressor/limiting (read any good books on decibel algo’s lately?) and of course the totally essential and “kewl” defining visualizations. 

   But we’ve got to find duplicates huh?  OK.  This is doable.  Not even rocket science and well within the reach of a startup mISV.  But the cost?

The cost is this.  It’s now bloated…

You’ve failed on the primary directive.  We know it’s primary because it was first in the list of “features” for the suggestion.  That primary was “Isn’t bloated.”

In summary.

1. Bloat is in the eye of the beholder.  Bloat to me is essential or acceptable to you.
2. Most people wouldn’t recognize quality Audio if it reached out, shoved a spoon down their ear and poured itself in.  That’s not just me saying that.  The proof of that is that people listen to MP3 files over the better fidelity CD or WAVE file and beyond those.
3. If you pull off the above you are competing with the big three freebies.  If you actually can do that, that is to say you are the incredible Jake the Peg with the extra leg, then somebody is going to clone you - or cut your extra leg off, whichever is easiest.  And they will. 

Next post I’m going to spoil other aspirations you might have, discuss strategies for picking fights with old ladies, holding intellectual political discussions with two year olds and…

Then again, I might just blog about my progress on MixAction….  

Scott Kane

Quote of the Day:
At the working man’s house hunger looks in but dares not enter.
–Benjamin Franklin

Related posts brought to you by Yet Another Related Posts Plugin.