I've been researching OS .NET ORM software offerings for a couple of days and my eyes are starting to bleed. It seems like ATOMSFramework is very well supported and has a good user base. I'm wondering if you can give me some of the reasons why ATOMS is better than the other solutions?
I'm especially impressed with iBatis, mostly because it's an official project of the Apache foundation. Also, it seems like it's pretty similar to ATOMS.
My specific project is pretty simple. We have an ASP.NET application that's grown up over the last year with direct ADO calls. We'd like to get out of the typing-for-the-fun-of-it business and use some sort of an ORM scheme -- but we'd like to do so with a minimum of learning curve (fat chance), in a staged phase-in instead of a complete rewrite.
I'd like to choose a foundation that we can move forward with on other projects. The problem isn't that there are too few choices, but too many.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Firstly, I hope the bleeding has stopped :-). Now on to the question of why the AtomsFramework is better than the others?
Damn good question and one that is a bit difficult to answer simply. You see, it all depends on what you value and what you are looking for in an O/R framework.
If you want to use one of the major O/R frameworks with plenty of developer interest/support, good documentation and the like then I would probably suggest that you look at either iBatis, NHibernate or DB4O (part of the mono project). Personally I've not used them and I don't really keep up to date with whats happening with them so I can't tell you their ins and outs.
All of us (O/R providers) will claim superior performance, ease of use, and all that hyperbole, but you really need to think about what you need to do.
Some frameworks require you to inherit from a base class to make objects persistable, others require the use of special syntaxes for object retrieval (NHibernate for instance) that are non-standard and take time to learn.
What the AtomsFramework tries to offer is flexibility. I've tried to provide multiple ways for things to be done - I want the framework to fit to the developer and not the other way around. Wether this has been done well enough only you can answer.
The framework itself is now pretty strong, but there are still areas that can be improved upon. Documentation is still a work in progress - the reference guide is up to date, but I need to still work on the user guides and rework the tutorials.
I'd also like to spend more time optimising performance to further reduce round tripping, and I don't currently cater for foreign keys in the database.
The only other issue you might want to consider is the language the framework is written in and how open the developers are to accepting patches and changes if you make any, or how likely feature requests will be actioned. The AtomsFramework is written in VB.NET and I'm very open to other people contributing changes or making suggestions. I can't speak for other projects.
I know this isn't the best answer in the world, but hopefully it gives you a better feel for how the AtomsFramework is positioned in relation to the other frameworks you have looked at.
My recommendation would be to download each of the "short list" choices and try them out on a simple test project. Maybe you can write a simple header/detail screen using your existing methods, then convert it to each of the possible framework choices. It might take a week or two to work through them all, but the time invested up front will be worthwhile.
- Richard.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for the informative response Richard! I hadn't heard of DB4O before, so I'll check that out.
Your recommendataion of trying out each framework on my short list is probably what I'll end up doing. No matter which one I end up using, I thank you for all the hard work you put into AtomsFramework -- both coding and supporting. It's guys like you who make the Open Source movement work.
Best regards,
John
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've been researching OS .NET ORM software offerings for a couple of days and my eyes are starting to bleed. It seems like ATOMSFramework is very well supported and has a good user base. I'm wondering if you can give me some of the reasons why ATOMS is better than the other solutions?
I'm especially impressed with iBatis, mostly because it's an official project of the Apache foundation. Also, it seems like it's pretty similar to ATOMS.
My specific project is pretty simple. We have an ASP.NET application that's grown up over the last year with direct ADO calls. We'd like to get out of the typing-for-the-fun-of-it business and use some sort of an ORM scheme -- but we'd like to do so with a minimum of learning curve (fat chance), in a staged phase-in instead of a complete rewrite.
I'd like to choose a foundation that we can move forward with on other projects. The problem isn't that there are too few choices, but too many.
Hi John,
Firstly, I hope the bleeding has stopped :-). Now on to the question of why the AtomsFramework is better than the others?
Damn good question and one that is a bit difficult to answer simply. You see, it all depends on what you value and what you are looking for in an O/R framework.
If you want to use one of the major O/R frameworks with plenty of developer interest/support, good documentation and the like then I would probably suggest that you look at either iBatis, NHibernate or DB4O (part of the mono project). Personally I've not used them and I don't really keep up to date with whats happening with them so I can't tell you their ins and outs.
All of us (O/R providers) will claim superior performance, ease of use, and all that hyperbole, but you really need to think about what you need to do.
Some frameworks require you to inherit from a base class to make objects persistable, others require the use of special syntaxes for object retrieval (NHibernate for instance) that are non-standard and take time to learn.
What the AtomsFramework tries to offer is flexibility. I've tried to provide multiple ways for things to be done - I want the framework to fit to the developer and not the other way around. Wether this has been done well enough only you can answer.
The framework itself is now pretty strong, but there are still areas that can be improved upon. Documentation is still a work in progress - the reference guide is up to date, but I need to still work on the user guides and rework the tutorials.
I'd also like to spend more time optimising performance to further reduce round tripping, and I don't currently cater for foreign keys in the database.
The only other issue you might want to consider is the language the framework is written in and how open the developers are to accepting patches and changes if you make any, or how likely feature requests will be actioned. The AtomsFramework is written in VB.NET and I'm very open to other people contributing changes or making suggestions. I can't speak for other projects.
I know this isn't the best answer in the world, but hopefully it gives you a better feel for how the AtomsFramework is positioned in relation to the other frameworks you have looked at.
My recommendation would be to download each of the "short list" choices and try them out on a simple test project. Maybe you can write a simple header/detail screen using your existing methods, then convert it to each of the possible framework choices. It might take a week or two to work through them all, but the time invested up front will be worthwhile.
- Richard.
Thanks for the informative response Richard! I hadn't heard of DB4O before, so I'll check that out.
Your recommendataion of trying out each framework on my short list is probably what I'll end up doing. No matter which one I end up using, I thank you for all the hard work you put into AtomsFramework -- both coding and supporting. It's guys like you who make the Open Source movement work.
Best regards,
John