Fog Creek Software
Discussion Board

I have done something stupid

I came from Bangalore, India to a college in US. In college in US, i developed a fantastic (according to my users) piece of freeware software.  It was a piece of tool designed to work with a database --  like TOAD or Embarcaderos Rapid SQL. I spent nearly 1.5-2 years part time on it. For me it was an oppurtunity to learn something and i did not care about money. Lot of people in Germany, North America etc. use and absolutely love the tool for what it does.

But unfortunately, my program works with a relatively unknown database. It does not work with Oracle I know feel i should have spent my time doing the same on Oracle.
The program is so damned good that it has features TOAD or Rapid SQL can only dream of. I could probably have made a lot of money from it had i written it for Oracle. Rapid SQL retails for 700 bucks. Could have undercut them like hell.

As i said, when i was developing it, i did not feel that i was wasting my time. Then , i thought that it would atleast  look good on my resume. I then discovered no company gives a damn to it.

Stupid, is it not?.  I now receive a letter asking me whether i have a tool like it for Oracle.

I have no illusions that i could have made millions out of it. But atleast, i would have been well known in the Oracle community and maybe make a couple of dollars too.

Had it been for a good well known database, i could have atleast demostrated it to Embarcadero or Quest and maybe got a job there continuing to work with it.

As i am not trying to sell or canvas the stuff,  the tools name or the database is not mentioned. Just looking for a shoulder to weep my sorry face on.

A Stupid Developer :-(
Monday, December 29, 2003

Can't you port it to a more salable database? Oracle is free for developers and IT professionals.

Bored Bystander
Monday, December 29, 2003

Can you divulge the name of the database it's based on? Someone who knows, say, ObjectStore in and out is very valuable to organizations who are tethered to ObjectStore.

Someone might be able to point you towards a company looking for that skill. Or, have you considered talking with your user community directly?

Mark Smith
Monday, December 29, 2003

I wouldn't call it a waste at all. Developing this tool would give you a general understanding of databases that few other graduates (or indeed, software developers in general) would have. The fact that your tool has great features and enthusiastic users are very positive things - it means you have a proven track record.

If you wanted to work with a database vendor (Oracle, Microsoft, etc) these potential employers will sit up and take notice of you when they see what you have done.

Alternatively, I'm sure you could use this experience to build an even better tool for a more mainstream database. You're in a position to build such a tool faster than most other people can, since you have experience to draw upon.

In short, you've created lots of great opportunities for yourself!

Monday, December 29, 2003

Why not develop a similar tool for Oracle's application server (OAS 10G, formerly 9iAS)?  Oracle's own web-based OEM(Enterprise Manager) for the apps server is absolutely awful, kludgy, slow and annoying.  A tool that is faster and presents more information with fewer clicks would certainly be welcome in the market.

There are dozens and dozens of alterable config files, and even more log files in OAS.  I'd love to have a tool that would let me quickly access and get to the end of these log files.

Market it w/ free monthly updates like the original TOAD, and maybe Quest Software will buy your app too. 

It's still not too late to develop your own TOAD-like tool.  The free version has many features disabled, and the pay version is expensive for individual developers to buy.

Monday, December 29, 2003

Port it and sell it.
Monday, December 29, 2003

Stupid Developer,

You need to stop beating yourself up, so first of all, stop calling yourself 'Stupid'. Remember, you're good enough, you're smart enough, and doggone it, people like you!

Stuart Smalley
Monday, December 29, 2003

Stupid Developer, boy are you stupid!

Monday, December 29, 2003

Go back to Bangalore and start a brand new absolutely fantastic and bombastic company. It would be easier for you to undercut TOAD and other products 'like hell' from there (land of opportunity). I hope I gave that badly needed shoulder to weep your face on.

Not that Stupid to fall for that stuff
Monday, December 29, 2003

This forum is becoming too elitist.

Unsuspecting programmers come here to talk, and they run into a wall of arrogant overachievers.

All the regular 120IQ people will leave, and the remaining handful of 180IQ scoffers will be left calling each other 'stupid' every time one of them can't remember Maxwell's equations by heart.
Monday, December 29, 2003

Don't beat yourself up too badly about this.  Look at it as an opportunity.  How?

Did you do any abstraction of the database layer when writing the tool?

If so, it may be relatively straightforward to port it.

If not, how long would it take to re-factor the database access methods out of the UI code to abstract them?

As someone (perhaps Bricklin or Ozzie) said, "great software takes ten years to build."

What year are you on?

dir at badblue com
Monday, December 29, 2003

>This forum is becoming too elitist.

Unsuspecting programmers come here to talk, and they run into a wall of arrogant overachievers.<

Word, Al, thas what I'm talkin' 'bout!

Monday, December 29, 2003

Off-thread: except for two posts (which I think were, possibly-lame, attempts at humor), there were no elitist or arrogant replies made.

Back to the original post, unless there is serious problems with abstraction in the original product, the suggestions make sense.

If you are just lamenting the missed/mismanaged opportunity of a market evidenced by the existance of popular products like TOAD and Rapid, you just have a boring and useless "hindsight is 20/20" "what might have been" problem. It is by no means clear that you would have had the other product's market (i.e. they could have stomped you).

It is very possible (though impossible to prove) that your current situation is a better opportunity: the other products prove that there is a market AND they display what are (presumedly) their weaknesses. Anyway, the current situation is really the only opportunity you have (i.e. the past is gone).

If you have the time, making your product work with Oracle would have, at least, the strong benefit of giving you Oracle experience. Even better, you might have a product that you could also sell profitably (maybe not for "millions" but for some real money).

"If you have been given lemons, make lemonade." Except that, if what you say is true, you have something better than lemons.

By the way, Rapid SQL works with many databases (natively with Sybase and Oracle, at least, and ODBC otherwise). This might be hard to duplicate.

Monday, December 29, 2003

It is not stupid that you gave it away for free.  If you charged for it, many of the people in Germany etc. who like it might not have bothered with it all.  Now you have gained their mindshare, and that can be a basis for developing and selling a 2.0 version with more features that works on other databases.

Linux would not have gone far if Linus charged money for the 1.0 version.

T. Norman
Monday, December 29, 2003

It is stupid that you gave it away for free. If it was genuinely useful, businesses would have paid for it eventually and you might have made money from it.

If Linus had charged for Linux 1.0, there would be 1,000's of programmers with a billion dollars or so. Linux fills the role of alternative OS and, had Linus not started it, someone else would.

Monday, December 29, 2003

Others have tried to create and sell alternative OS. Look what happened to OS/2 and BeOS.

T. Norman
Monday, December 29, 2003

"All the regular 120IQ people will leave, and the remaining handful of 180IQ scoffers will be left calling each other 'stupid' every time one of them can't remember Maxwell's equations by heart."

Yikes!  You just gave me a nightmarish flashback of the college Physical Chemistry prof whose testing philosophy was that all our tests should be closed book, closed notes because the students should just memorize Maxwell's equations and derive everything on the fly during the test.  Man, I hated that class.

Matt Latourette
Monday, December 29, 2003

Hmmm, I very much liked my high-school physics class, whose teacher who was a big believer in having students be able to derive things from first principles. That was useful because instead of simply memorizing equations and plugging in numbers, we were forced to analyze problems and determine what principles applied before getting into any specifics. That was a great experience, and excellent practice for the inevitable problems that didn't parallel anything you'd seen before, but were still quite solvable if you understood the fundamentals and could build a case from that.

Of course, I never took p-chem so I have no idea if this analogy is any good.

John C.
Monday, December 29, 2003

Thank you everyone for your comments .
I appreciate it very much.

What i will do is to start porting it to Oracle 10 once the release is out via OLE DB. Might take me a couple of years. But atleast, i will get to know another database.

I deeply appreciate all  responses ;-). Its really encouraging.

A Stupid Developer :-(
Monday, December 29, 2003

Stupid is as stupid does!

Run Forrest Run
Monday, December 29, 2003

Did someone say Maxwell's equations? Know them by heart. Always preferred the differential form. I'd write them out, but they're impossible to put into html. BTW, I agree that they're pretty much all you need to know - the rest can be derived from them.

The original poster is only stupid if he doesn't port to a popular database. He claims he could make millions. Let's assume that it's 3 million and takes 3 months to do the port.

That's a pay rate of:

1 million dollars per month
or $45 454 per work day (approx, taking weekends off)
or $5681 per work hour (assuming he works 8 hour days)

Plus he gets to be his own boss.

Monday, December 29, 2003


i did NOT say i could make millions. Please read the post

have no illusions that i could have made millions out of it.>>

A Stupid Developer :-(
Tuesday, December 30, 2003

Unless the database engine you worked with is entirely a different shape I can't see that it would be a huge problem to port it.  When you do port it, don't port it simply to Oracle create an interface that allows you to port it to any database engine.

That argues for an ODBC/JDBC type of interface but that needn't be the case.  Take a look at the sources for the Perl DB connectors and the Python equivalents that should give you a running start on what you need to do.

Don't consider porting the front end of the application only the lowest level of database call, hide all the details.  If its a success then in subsequent versions your can walk higher up your application replacing and augmenting to get the features of other databases into your code.

Simon Lucy
Tuesday, December 30, 2003

Sorry. My mistake. I saw the dollar signs and lost my mind.

Tuesday, December 30, 2003

Simon Lucy: actually connecting to the database and retrieving records from it is the easy part of abstracting the database layer. The hard part is making sure the SQL does the same thing.

I know many databases vary significantly between themselves in the syntax of the SQL queries. While the ANSI SQL '99 standard defines the basic syntax, the built-in functions and many other conventions vary a great deal.

Shlomi Fish
Tuesday, December 30, 2003

This is true, but I wouldn't expect such a tool to use builtins other than simple ones such as sum().  If you limit yourself to something like a generic ODBC layer (without passthru), then you should get back the same results for the same query.

Simon Lucy
Tuesday, December 30, 2003

"Hmmm, I very much liked my high-school physics class, whose teacher who was a big believer in having students be able to derive things from first principles."

I agree that it's useful to know how to derive things.  The issue was that it was totally inappropriate to derive everything on the fly during a time-limited test situation, as evidenced by the scores people got on his exams.  Tests were multiple choice with no partial credit for your work and the tests were so difficult that a score of 50% would've gotten you an A grade.  I changed majors to comp. sci. after I had already signed up for the second term of this course and decided that I no longer cared about this class in any way other than minimizing it's impact on my overall GPA, but I couldn't get another class to replace it and couldn't drop it so I just showed up for the exams only.  I simply crossed out any answers that looked obviously wrong and used the random number generator on my HP 48GX calculator to decide among the remaining options.  I got a passing grade in the class.  Any prof that thinks exams like that are valid has never heard the term signal-to-noise ratio.

Matt Latourette
Tuesday, December 30, 2003

Stupid Developer,

You sound brilliant to me.

Brian R.
Sunday, January 4, 2004

*  Recent Topics

*  Fog Creek Home