I've got to throw in my two cents...

First, like all the others have stated, LLBL Gen Pro is a great product at a great price with great support!

Second, like Frans, I can't stand complicated licensing schemes, or support contracts, or anything else that muddies the water. I have a day job that doesn't require LLBL Gen Pro, but I am working on a project at home in my spare time that I hope to market later. I could have built my data access layer myself (again) but finally decided to give LLBL Gen Pro a try since I had seen Frans and Thomas hawking their products so much at the asp.net forums (Actually, I paid more for LLBL Gen Pro because I bought CodeSmith first after reading that very long thread architecture thread. Then, I reread the thread a few weeks later and decided to drink the ORM Kool-Aid ). Even though I don't really have the financial cushion (three kids and a stay-at-home wife) to spend on products for something that may or may not pan out, the price for LLBL Gen Pro was just right. Had it been priced higher or had some crazy pricing matrix, I wouldn't (couldn't) have bought it. But since I have, I am yet another Solutions Design evangelist.

Third, I understand that all of you who think that LLBL Gen Pro is underpriced are just wanting to help Frans, but please quit telling him to raise the price and quit improving the product . I know I've already paid for it, but I want future people to run across the same great combination of features + support + price that we did. Plus, I know I will upgrade to .Net 2.0 this fall and don't look forward to having to spend more money that I don't have.

Thanks Otis for a great product. And thanks to forum members for a great community.

Matt

Bah, I don't want to re-hash things I've said previously, but I do want to comment on a few posts.

Otis wrote:

the market isn't growing fast, but keeping steady

Small pond, big fish. Grow the pond by introducing your product in a more friendly way. Eventually you'll be known by every developer out there - seriously, it can happen. The only way to get the straglers is to help them learn why they'd be helping themselves. And by straglers I mean the 80% of the developers who don't consider themselves good enough developers to learn your product. I know. I work with these people. Make the product easier to understand for newbies. ObjectSpaces meant that Microsoft was going to do the heavy lifting of introducing the concept to developers. Now it's your job, and the rewards are tremendous. Consider creating a new website, just for ORM education, or consider hosting forums, affiliated with CodeProject for example, just on that topic. Be an evangelist outside the borders of your own product. Consider certifying people on your product, or ORM itself; these people often become evangelists. Grow the pond. I'll say no more.

jeffreygg wrote:

...and they provide a certain amount of value, it's hard to ask for a substantial amount of money in return for a hard to justify increased level of value.

...unless they understand up front what the benefits will be. Anyone who has used ORMs, or fully understands the concept knows the value here, even though a portion is supplied by the end-user. The very topic here is that those of us who understand the value recognize that it's worth more, even though we understand we're contributing as well. It's a job of education first, pricing second. People will pay more if they understand as we do. They must be made to understand before the purchase.

mattsmith321 wrote:

...but I am working on a project at home in my spare time that I hope to market later.

Every good coder I've ever known has a project going on the side. I've hired dozens of developers in past jobs, and I always ask the question during an interview, "What are you working on at home?" If they say they aren't, their either a liar, or are not worth knowing. Good for you. You are more valuable than you know.

jeffreygg wrote:

You know, the other thought I had was that my points about holding features over for new versions only makes sense in a saturated, or nearly saturated market. If the rate of growth is holding steady or increasing, and there's still a large market to be obtained, then it makes sense to get the most out of the existing investment, ie, the current version. Why build a new version when you can just add a feature to the existing version, still gain additional customers AND make your existing customers even more happy. That makes sense. You won't get any additional money from your existing customers, but you will steadily gain customers for the first version, which means even there will be even more potential revenue when you do decide to offer a new release as you now have a larger installed base from which to attract upgrade customers.

It is kind of a difficult decision, IMHO: add the feature to the current version, or mark the current version + new feature the 'new version'.

We don't advertise, at least not anymore. It didn't bring any extra revenue, online banners are not that effective really. What we do see though is that mouth to mouth advertising: developers telling other developers what to use, is paying of a lot. To achieve that, current customers have to become really pleased So for us the difficult decision is made in favor of current version + new features == current version .

mattsmith321 wrote:

Third, I understand that all of you who think that LLBL Gen Pro is underpriced are just wanting to help Frans, but please quit telling him to raise the price and quit improving the product . I know I've already paid for it, but I want future people to run across the same great combination of features + support + price that we did. Plus, I know I will upgrade to .Net 2.0 this fall and don't look forward to having to spend more money that I don't have.

The current codebase won't be dropped, it will be supported for at least another year. Just not more new features but bugfixes will be released for the current codebase for at least a year after 2.0 is released: not everyone upgrades to 2.0 of .NET, so an alternative has to be available and that alternative is fully supported.

swallace wrote:

Bah, I don't want to re-hash things I've said previously, but I do want to comment on a few posts.

Otis wrote:

the market isn't growing fast, but keeping steady

Small pond, big fish. Grow the pond by introducing your product in a more friendly way. Eventually you'll be known by every developer out there - seriously, it can happen. The only way to get the straglers is to help them learn why they'd be helping themselves. And by straglers I mean the 80% of the developers who don't consider themselves good enough developers to learn your product. I know. I work with these people. Make the product easier to understand for newbies. ObjectSpaces meant that Microsoft was going to do the heavy lifting of introducing the concept to developers. Now it's your job, and the rewards are tremendous. Consider creating a new website, just for ORM education, or consider hosting forums, affiliated with CodeProject for example, just on that topic. Be an evangelist outside the borders of your own product. Consider certifying people on your product, or ORM itself; these people often become evangelists. Grow the pond. I'll say no more.

Thanks, Scott . It's always a pleasure reading your POV, as you, and others, bring with your POV things to the table I would never think about I'll sincerily consider your suggestions. The only thing that makes me hesitate a bit is that LLBLGen Pro is not a pure O/R mapper, which makes the advocacy I might do a bit off-base. (This is political blabla and no-one should care, but inside the O/R mapper community there are unfortunately a couple of camps with very vocal advocates... and they don't like what the others have to say. This doesn't help much, but apparently they don't care). But there must be a way

The only thing that makes me hesitate a bit is that LLBLGen Pro is not a pure O/R mapper

Why not? Explain more please...

wayne wrote:

The only thing that makes me hesitate a bit is that LLBLGen Pro is not a pure O/R mapper

Why not? Explain more please...

'Pure O/R mappers' (there is no law for this, but the common understanding of the term is this) work from the domain model towards the database, have plain classes as base, support inheritance and generate the database for the object model, not the other way around. Pure O/R mappers also focus solely on objects, not the relational model. This means that nice things like typed lists, views etc. are things which are not used.

Otis wrote:

It's always a pleasure reading your POV, as you, and others, bring with your POV things to the table I would never think about

Thanks for seeing it for what it is, constructive and meant to help. You're a pleasure to do business with.

Here's a link to a nice interview with Josh Clark, head of Global Moxie and creator of Big Medium: http://www.gadgetopia.com/2005/03/02/InterviewWithJoshClark.html

I've never heard of him, his company, or his product. He does however have some interesting things to say about running his own small business. He seems to have similar philosophies to Frans and it sounds like he probably has a fanatic user-base just like LLBL Gen Pro. That almost makes me want to try his product even though I have no use for it.

No point really. Just thought the article and this thread were somewhat similar.

Enjoy!

mattsmith321 wrote:

Here's a link to a nice interview with Josh Clark, head of Global Moxie and creator of Big Medium: http://www.gadgetopia.com/2005/03/02/InterviewWithJoshClark.html

I've never heard of him, his company, or his product. He does however have some interesting things to say about running his own small business. He seems to have similar philosophies to Frans and it sounds like he probably has a fanatic user-base just like LLBL Gen Pro. That almost makes me want to try his product even though I have no use for it.

Thanks for that great link! Man, it read as if I was the one who was interviewed...

Otis wrote:

wayne wrote:

The only thing that makes me hesitate a bit is that LLBLGen Pro is not a pure O/R mapper

Why not? Explain more please...

'Pure O/R mappers' (there is no law for this, but the common understanding of the term is this) work from the domain model towards the database, have plain classes as base, support inheritance and generate the database for the object model, not the other way around. Pure O/R mappers also focus solely on objects, not the relational model. This means that nice things like typed lists, views etc. are things which are not used.

I think you just described an OPF rather than O/R mapper.

TObject wrote:

Otis wrote:

wayne wrote:

The only thing that makes me hesitate a bit is that LLBLGen Pro is not a pure O/R mapper

Why not? Explain more please...

'Pure O/R mappers' (there is no law for this, but the common understanding of the term is this) work from the domain model towards the database, have plain classes as base, support inheritance and generate the database for the object model, not the other way around. Pure O/R mappers also focus solely on objects, not the relational model. This means that nice things like typed lists, views etc. are things which are not used.

I think you just described an OPF rather than O/R mapper.

OPF as in Object Persistence Framework? But what's the difference between Object persistence framework and O/R mapper? (just curious)