Revolution and the Web, feedback wanted, Part 1 of 3

[Brian Yennie]

Just as a tie-in here, I don't mean to bash Morfik. It looks like an  
interesting technology. But as a startup business owner especially,  
you can't ignore certain things, like:

1) High licensing fees (even if they are discounted) for an unproven  
product
2) 6 years in development and no successful case studies outside of  
the company
3) Litigious activity (apparently they are trying to patent Java =>  
Javascript compilation as if they invented it)
4) Bad review (the only one I could find): http://mudabone.com/aietc/? 
p=346
5) Google offers something similar which is actually running  
massively used services (Google Web Toolkit)

> It's not really "retroactive".  You owe the seat fee, as of the date
> when you start making money, subject to whatever actual deal you work
> out with them.  They do state that they will work these terms out on a
> case by case basis with each startup.  And yes, I would agree to this,
> because I have zero risk in trying it out.  On top of that I can just
> take the Express version for FREE until I'm confident that it's going
> to work for me, then go Pro.

Fair enough on the Express version. But if the express version if  
sufficient, what's the value of the startup deal? How about this  
analogy - it's basically a credit line. Eventually you owe the  
balance. If the license is perpetual, then it kind of is retroactive  
if you're paying the same money you would have upfront. Either way,  
definitely not free. By the time you have any money, you're still  
spending it on licenses. Credit lines can be good and bad, that's  
all. And since this arrangement is so unconventional, it may make  
business owners nervous.

> Worrying about the 1.0 version is a little odd when discussing a
> business decision broadly.

No it's not. Have you ever run a startup, with your personal finances  
at risk? You don't think it's a major factor to risk your livelihood  
on a product that is unproven in the market? Maybe you evaluate the  
product and are satisfied, but the 1.0 factor has got to be  
considered when your business is on the line.

> First of all, the startup that is looking
> at this deal isn't even at .1 yet.  Second, the price isn't $5,000.
> The 1.0 price is $1,748 until February or so, if I remember the dates
> correctly.

I know - but it's just a promo price, so I don't know where it lies  
in the big picture with this product. And $1,748 is still a  
substantive professional license. The fact that the startup company  
doesn't have a product at all just hammers home the fact that they  
are at great risk already. Startups don't necessarily want to partner  
up with startup technology and amplify their already high risks.

> Thirdly, what would you say if they took $7,000 in stock
> options per seat instead of cash when you start generating revenue?

I'd say, are you kidding me? I have to give away equity in my  
business to use a developer tool that exports industry standard code?  
That's insane!

> Would it be a good deal then?  Finally, arguing that any pricing or
> any product is a worry because it is 1.0 does not make the product or
> the pricing a problem in even the medium term.  If this was the deal
> and the product was 1.5 (which one would expect it will be sometime
> next year), would you still think it's a problem?

My comments in regards to 1.0 are key for startups. More established  
businesses may be better prepared to counter the risk. But yeah,  
invested thousands of dollars per seat in professional licenses for a  
1.0 product is a very risky move. I'm not saying it's fair to the  
vendor, especially if it's a great product. But it's reality.

There's no guarantee that next year will yield 1.5. In fact, anyone  
who's dealt with startup vendors over the years knows there is a good  
chance that there won't be any new versions if they don't start  
selling licenses ASAP. The fact that they are offering heavily  
discounted licenses tells me that they aren't out of the woods yet.

> Yes, they are bundling in software that doesn't cost them anything
> (which means no license fee for SQL Server or Oracle).
> The thing that
> interests me, though is that they are hiding these components from me,
> and they are making it easy to detach them (if you decide to go Pro).
> I don't have to know how to configure Firebird or Apache, or write a
> bunch of Perl or PHP.  I design my application the way I would if I
> were using any other RAD IDE, and when I compile it, Firebird and
> Apache are built in.

The fact that Morfik apparently outputs a custom .exe file with the  
database and web server embedded sounds truly awful to me. So you're  
saying I can only run this on Windows, it won't scale, it won't  
conform to any normal Apache or Firebird documentation, I can't  
upgrade the web server or database without rebuilding my app, and I  
can't run this on 99% of the web hosts out there?

> Yes, you can do this in a variety of ways.  I have not seen any
> implementations that _efficiently_ use bandwidth that use anything
> OTHER THAN XML to embed the data.  If you can do it, great.  Maybe
> we'll get some of the frameworks tweaked to make life less
> <taghappy/>.

Yahoo web services, for starters, supports both XML and JSON. What  
implementations have you seen that are using something other than XML  
to be less efficient? If you're referring to just sending back badly  
written HTML snippets then yes, that's worse.

- Brian