Your Blog post of Dec 23

Posts   
 
    
pilotboba
User
Posts: 434
Joined: 05-Aug-2005
# Posted on: 23-Dec-2005 16:17:52   

Frans,

In your blog post you state:

"Now it would be great if Microsoft for once would start writing some proper do-cu-men-ta-tion for design time databinding supporting code."

Are you kidding, this would destroy the market for third party books that explain this stuff.

BOb

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39800
Joined: 17-Aug-2003
# Posted on: 23-Dec-2005 17:22:00   

heh simple_smile You're writing one, I pressume? simple_smile

However, if I have to buy a book to learn how things work in vs.net, I think we've reached the end of the direction they're going right now. At the moment, I lose a lot of time by simply fighting with VS.NET 2005, while it should be a couple of hours max to write out all the code required...

(edit): most time is lost by trying to make vs.net 2005 accept tne entitycollection again as a toolbox item. removing obj files, recompiling, sometimes it works, sometimes it doesn't. Really sad, as it is stupidity all over. "Can't create toolbox item blabla", is what I get, no explanation why, and when I attach a debugger to the toolboxitem code, nothing gets called, or the error is somewhere else. disappointed

Not something to get all warm and excited about. disappointed . On the upside, my first EntityView2<CustomerEntity>()(entitycollection) worked in a databinding environment, so there is some progress finally wink

Frans Bouma | Lead developer LLBLGen Pro
pilotboba
User
Posts: 434
Joined: 05-Aug-2005
# Posted on: 23-Dec-2005 20:23:01   

Otis wrote:

heh simple_smile You're writing one, I pressume? simple_smile

Ha, not at the moment. My first and last book was on a third party Visual FoxPro framework which took a year to write all 180 pages.

But, I have considered writting an LLBLGen book... but I'm not sure how big the market for that would be. I think doing niche books is a better choice since while your market is smaller, you also have less competition for readers. Also, not sure who would want to publish it, perhaps I could do it as an eBook.

Otis wrote:

However, if I have to buy a book to learn how things work in vs.net, I think we've reached the end of the direction they're going right now. At the moment, I lose a lot of time by simply fighting with VS.NET 2005, while it should be a couple of hours max to write out all the code required...

Well, you shouldn't have to read a book for the "main stream" stuff, and creating a bindable custom collection seems pretty main stream to me.

Also, that's why you make the "big bucks" (why people buy your stuff) because you do all the hard work and we just stand on the giants shoulders.

BOb

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39800
Joined: 17-Aug-2003
# Posted on: 29-Dec-2005 11:28:37   

pilotboba wrote:

Otis wrote:

heh simple_smile You're writing one, I pressume? simple_smile

Ha, not at the moment. My first and last book was on a third party Visual FoxPro framework which took a year to write all 180 pages.

cool simple_smile . I thought about writing a book but I simply can't find the motivation to write that amount of text. I saw the manual is now over 300 printed pages already, but a book needs so much more... I have deep respect for everyone who can finish writing a book.

But, I have considered writting an LLBLGen book... but I'm not sure how big the market for that would be. I think doing niche books is a better choice since while your market is smaller, you also have less competition for readers. Also, not sure who would want to publish it, perhaps I could do it as an eBook.

Would be great! It still would be a niche book indeed, so don't expect the money to buy a ferrari wink . The thing is that computer books in general hardly sell over 10,000 copies, even if they're top of the list. I was very surprised when I hearded that, I thought the top sellers would sell over 100,000 copies easily.

Otis wrote:

However, if I have to buy a book to learn how things work in vs.net, I think we've reached the end of the direction they're going right now. At the moment, I lose a lot of time by simply fighting with VS.NET 2005, while it should be a couple of hours max to write out all the code required...

Well, you shouldn't have to read a book for the "main stream" stuff, and creating a bindable custom collection seems pretty main stream to me.

Well, finding ITypedList info was tough a couple of years ago. Ironically, the article written by one of teh ORM.NET developers about ITypedList learned me what I had to do wink .

After the binding is setup, it's pretty straight forward. What's a challenge is design time databinding. For example: how do you retrieve all factory types from the current solution? At the moment I use a trick, as the EntityCollection is in a project, and thus I can grab the assembly it's in and then find the factories. But if I move the ENtityCollection to the base library, I can't anymore. Very recently I stumbled accross a new interface in .NET 2.0, ITypeDiscoveryService, which is retrievable through the Site instance, but it's as usual with this kind of stuff, hardly documented.

Also, that's why you make the "big bucks" (why people buy your stuff) because you do all the hard work and we just stand on the giants shoulders.

True simple_smile . You won't hear me complaining wink It's just that I wished MS would put as much effort in documenting the harder parts of their API as what they put into documenting the general parts more people would use.

Frans Bouma | Lead developer LLBLGen Pro
pilotboba
User
Posts: 434
Joined: 05-Aug-2005
# Posted on: 03-Jan-2006 19:52:40   

Otis wrote:

Would be great! It still would be a niche book indeed, so don't expect the money to buy a ferrari wink . The thing is that computer books in general hardly sell over 10,000 copies, even if they're top of the list. I was very surprised when I hearded that, I thought the top sellers would sell over 100,000 copies easily.

Yea, I think my last royalty deposit was $19. I'm pretty surprised that one or two a month still sell.

I think the 10,000 for best seller mark is a sad state of affairs and it shows that alot of folks that are "programmers" are just hackers rather than craftsman. If I had more money and time I would read alot more technical books.

Perhaps this is why so much software is over budget, late and of low quality, eh?

BOb

Otis avatar
Otis
LLBLGen Pro Team
Posts: 39800
Joined: 17-Aug-2003
# Posted on: 04-Jan-2006 12:29:52   

pilotboba wrote:

Otis wrote:

Would be great! It still would be a niche book indeed, so don't expect the money to buy a ferrari wink . The thing is that computer books in general hardly sell over 10,000 copies, even if they're top of the list. I was very surprised when I hearded that, I thought the top sellers would sell over 100,000 copies easily.

Yea, I think my last royalty deposit was $19. I'm pretty surprised that one or two a month still sell.

that's hardly enough to get a decent meal at a restaurant!

I think the 10,000 for best seller mark is a sad state of affairs and it shows that alot of folks that are "programmers" are just hackers rather than craftsman. If I had more money and time I would read alot more technical books. Perhaps this is why so much software is over budget, late and of low quality, eh?

Well, I also think it's the fault of the publishers. As I was a technical reviewer of Sahil Malik's ado.net 2.0 book for Apress, I know hte schedules those publishers are pushing and it's simply undoable to get great content under those conditions. Another thing is that publishers publish more books which are tied to a (hyped) technology which is 'hot' than books which are more generic and focussed on real knowledge applyable everywhere in CS.

I mean: take a random 'learn C#' book. There are very few which solely discuss C# as a language and go deep inside the language. Most of them (if not all) discuss the elements you can do and then spend at least half of the book on .NET software writing. If I take Kernighan-Richie: 'The C programming language' (2nd edition wink ) I can only see C-language oriented chapters. That's why the book is good (IMHO) and worth keeping. Then you'll pick a book from the shelve to read back a chapter how things were meant to be done because you KNOW everything there is to know about the language is in that book. Now not everything is in there, as half of the books are spend on writing stuff in .NET which has no place in that book.

I must confess that I hardly buy technical books these days. One of the biggest caveats I find with physical paper books is that there's no search function wink . The digital ones are also horrible to read because of the adobe software.

I also get more and more the idea that people simply write technical books these days to get attention so they get speaker jobs at conferences. (which is sad, because no-one would then focus on really teaching people things, just jump on the latest hype train)...

Frans Bouma | Lead developer LLBLGen Pro
Posts: 497
Joined: 08-Apr-2004
# Posted on: 04-Jan-2006 13:04:58   

Tip for Adobe Acrobat reader users - do a google search for "foxit reader", and use that instead simple_smile