One thing I noted differently about the LLBL forum after being absent for sol long is that its not as active as it was one to one and half years back.

I dont know if this is a bad thing or a good thing. Maybe because the product has matured so well and us its users have also done so.

What I hope for (and I know this one will make Frans happy) is to see LLBL's user base grow. I just hope the diminished activity is not a sign of decrease in the growth of the product's popularity!!

When I look at the # of posts per month, I see that in 2006, it's on average about 1585 per month, in 2007 about the same (1622) and in 2008, it's about 1080, though just 2 months going. So 40-50 posts a day. I think that's pretty active, considering the forums of competitors, also free solutions, are all less active.

What we've changed in the last 1.5 year is that we've added a faq here, and the more questions are asked the more answers the search will provide. We also have more examples now.

Our userbase increases in a faster rate than 1.5 year back so I don't think it's that . I think the number of examples which is now bigger, the bigger manual, way more answers to find on this forum and the FAQ which is now also in this forum (so pops up in the search) makes a lot of difference. After all, this is a Q&A forum most of the time: people as a question, get an answer, move on.

We definitely don't see a decline in interest in LLBLGen Pro, on the contrary. Of course there are more competitors than 2 years ago, but at the same time, the market got more mature, so the few players which are on everyone's shortlist got more mature (we're on that list, don't worry) and with that, the players who just started have a hard time to catch up and get the attention of the people interested in o/r mapping on .net.

(edit) Thanks for your concern btw. Don't worry about a decline, it's not happening, on the contrary. Also, with our Linq support just around the corner, I think we're set for a sunny 2008

Thanks FRANS... seems you will always be our rock.

I have to admit I cant wait to put my hands on LINQ2LLBL

omar wrote:

Thanks FRANS... seems you will always be our rock.

I hope so hehe

I have to admit I cant wait to put my hands on LINQ2LLBL

I hope it will meet the expectations people have. Yesterday I've finished prefetch path development, today I hope to add fast hierarchical dynamic list projections (with anonymous typed objects in lists), at least a start with the code .

(eg:

var q = from c in metaData.Customer
            where c.Country == "Kuwait"
            select new {
                             c.CustomerID,
                             Orders = (from o in metaData.Order
                                             where o.CustomerID==c.CustomerID
                                             select new { o.OrderID, o.OrderDate})
                              };

Linq to sql produces (1 + #customers) amount of queries, as it simply places the query for orders as an IEnumerable in the anonymous typed object with customer data. When you access that (and the engine does that directly when populating the list), it executes the query, but just for that customer...

Of course, this is a simpler form of prefetch paths: it can use the same code, at least in many occasions. There are situations where it can't do this, so there I've to find a solution for but for the ones which are optimizable, i.e. a nested set with a correlation relation between the nested set (orders) and the parent (customers) could be done with 1 query per graph node, in this case: 2 queries: one for the customers and one for all the orders of these customers

Otis wrote:

What we've changed in the last 1.5 year is that we've added a faq here, and the more questions are asked the more answers the search will provide. We also have more examples now.

Those are good reasons.

I've also wondered if a lot of the questions end up in the Helpdesk forum.

In addition, back in the day, Frans was everywhere stirring up trouble. In other forums or on other blogs. Remember the really long threads about stored procs versus dynamic SQL (with Rob) and the code generators versus ORM (with Eric, Thomas and Paul)? Those discussions generated a lot of buzz and interest in what was going on over here.

mattsmith321 wrote:

Otis wrote:

What we've changed in the last 1.5 year is that we've added a faq here, and the more questions are asked the more answers the search will provide. We also have more examples now.

Those are good reasons.

I've also wondered if a lot of the questions end up in the Helpdesk forum.

I think most questions end up in the normal forums. I did a SQL query on the database with a couple of aggregates to get the numbers above so that's overall on the forums, not just the public ones.

In addition, back in the day, Frans was everywhere stirring up trouble. In other forums or on other blogs. Remember the really long threads about stored procs versus dynamic SQL (with Rob) and the code generators versus ORM (with Eric, Thomas and Paul)? Those discussions generated a lot of buzz and interest in what was going on over here.

Heh, yeah those were the days . Those discussions took a LOT of time, and looking back, it's been worth it but still if I have to do it again I think I'll pass. (though that blogentry about procs are bad has been read hundreds of thousands of times already, really amazing).

Ironically that Eric has dissapeared apparently (codesmith, is that still going on?), Thomas has moved on to a CMS company and Paul left the o/r mapper business.

Otis wrote:

Heh, yeah those were the days . Those discussions took a LOT of time, and looking back, it's been worth it but still if I have to do it again I think I'll pass. (though that blogentry about procs are bad has been read hundreds of thousands of times already, really amazing).

I refer people to that discussion fairly often when they start quoting the standard 'stored procs are faster' fallacies.

Otis wrote:

Ironically that Eric has dissapeared apparently (codesmith, is that still going on?), Thomas has moved on to a CMS company and Paul left the o/r mapper business.

Even more ironic is the fact that Eric's CodeSmith has joined up with Rob's Telligent:

Recently, CodeSmith Tools, LLC was formed as a partnership between Eric J. Smith and Telligent Systems as the first step to fulfilling our vision of becoming a great developer tools company.

from http://www.codesmithtools.com/aboutus.aspx

Maybe everyone's going to Frans' blog instead? btw, I'm loving all the linq posts!

mattsmith321 wrote:

Otis wrote:

Heh, yeah those were the days . Those discussions took a LOT of time, and looking back, it's been worth it but still if I have to do it again I think I'll pass. (though that blogentry about procs are bad has been read hundreds of thousands of times already, really amazing).

I refer people to that discussion fairly often when they start quoting the standard 'stored procs are faster' fallacies.

Heh

I think it's better to refer to this summary post which also contains the link to my theserverside.net post about this which I find better than the stored procs are bad, mkay post

Otis wrote:

Ironically that Eric has dissapeared apparently (codesmith, is that still going on?), Thomas has moved on to a CMS company and Paul left the o/r mapper business.

Even more ironic is the fact that Eric's CodeSmith has joined up with Rob's Telligent:

Recently, CodeSmith Tools, LLC was formed as a partnership between Eric J. Smith and Telligent Systems as the first step to fulfilling our vision of becoming a great developer tools company.

from http://www.codesmithtools.com/aboutus.aspx

Yes I know, still there aren't much results coming from that corner. Except perhaps that they won the battle with mygeneration

luciusism wrote:

Maybe everyone's going to Frans' blog instead? btw, I'm loving all the linq posts!

Thanks! One more coming up hopefully today!

Otis wrote:

Yes I know, still there aren't much results coming from that corner. Except perhaps that they won the battle with mygeneration

Did they? Why do you say that? Is MyGeneration abandoned? Didn't those guys go on to do a commercial code gen tool?

Of course, there is a new code gen tool on the market now in Code Breeze. Then there is Kath Dollard working on her new stuff that uses VB's XML litteral's support for templates... and all here stuff is going to be open source.

BOb

PS: Yes, I know your not in the code gen market, but I think alot of people see your production that way.