DevTracker updates
So since the last post, I broke my demo! Not happy about it, but it signified a good time to move on. I changed the database substantially and then the auto-generated code wouldn't work any longer no matter what I tried. After about 15 minutes I figured it was time to revamp the database some more and then move onto iteration 1 of the Silverlight version of DevTracker.
So, I redid many of the things I did the last project with the full DevTracker database. I had integrated my own user management with the built in login functionality and then broke it, but that is bonus stuff since the first version of the project is for myself so I don't really need multiple users yet. I am building it with that in mind however. But I'm almost at the point where I can actually use it in the development process of itself. That's right! This software is already beginning to track and plan its own completion. I can create and manage projects and phases, and I can add tasks, but currently I can't see tasks or alter them, just added them now and I plan on something big, new and exciting for the tasks. Once I can view them and mark them as completed I will have enough to really start rolling.
So, from the last post, even though I broke out of the demo before completing any of my bullet points, I have already done 2 of them in the software. As stated, I integrated the existing login functionality with my own database to a degree and I have some more complex sources of data that allow me to select related entities from a combo box (doing this for my tasks).
So, now that I'm moving ahead with my Silverlight apps, and I still am not tired, I thought I would also throw out my updated thoughts on Silverlight and EMF. I'm loving it now!!! There are still times where things are harder to debug or I have to reach out to the internet to figure it out, but I think this works better than my original plans for the my potential side business here in Barrie. These 2 frameworks allow me to build rich looking applications with rich functionality very quickly. Yes, there are still downsides. After my experience with tweaking the database I've decided that the entity model framework isn't a good choice if the database will be changing a lot. Fortunately I don't see that being an issue with quick turn-key projects hosted on the web. Database changes caused me no end of head aches in my demo app, and I saw a number of deficiencies versus what I was used to with Fluent nHibernate.
Ultimately, if I start a business and I'm getting a lot of steam behind a certain project or concept, I would probably port it back into WCF with nHibernate if Silverlight was still requirement or go right back to WPF and nHibernate if not. The great thing about the exposure to Silverlight, is that Silverlight is a neutered version of WPF, so going in that direction should be easier than the other way around.
Anyway, thinking about a potential business is still a ways off. After DevTracker... the next project it will track is the Silverlight version of Veronica's software, provided my reporting solution will work in Silverlight. I spent enough time digging up that solution that I WILL scrap Silverlight for that side of things if I can't make it work. Then after that will be a modular but basic business accounting package, proper tracking of ledgers is something Veronica, myself and probably many of my currently fictional future customers would be interested in. And then after that... I have 1 or 2 projects I've always been tempted to start for fun. Both could make money, so they may be the platform for my future enterprises.
Anyway, tired or not, getting late and I need some sort of sleep if I plan on working 8 hours tomorrow.
So, I redid many of the things I did the last project with the full DevTracker database. I had integrated my own user management with the built in login functionality and then broke it, but that is bonus stuff since the first version of the project is for myself so I don't really need multiple users yet. I am building it with that in mind however. But I'm almost at the point where I can actually use it in the development process of itself. That's right! This software is already beginning to track and plan its own completion. I can create and manage projects and phases, and I can add tasks, but currently I can't see tasks or alter them, just added them now and I plan on something big, new and exciting for the tasks. Once I can view them and mark them as completed I will have enough to really start rolling.
So, from the last post, even though I broke out of the demo before completing any of my bullet points, I have already done 2 of them in the software. As stated, I integrated the existing login functionality with my own database to a degree and I have some more complex sources of data that allow me to select related entities from a combo box (doing this for my tasks).
So, now that I'm moving ahead with my Silverlight apps, and I still am not tired, I thought I would also throw out my updated thoughts on Silverlight and EMF. I'm loving it now!!! There are still times where things are harder to debug or I have to reach out to the internet to figure it out, but I think this works better than my original plans for the my potential side business here in Barrie. These 2 frameworks allow me to build rich looking applications with rich functionality very quickly. Yes, there are still downsides. After my experience with tweaking the database I've decided that the entity model framework isn't a good choice if the database will be changing a lot. Fortunately I don't see that being an issue with quick turn-key projects hosted on the web. Database changes caused me no end of head aches in my demo app, and I saw a number of deficiencies versus what I was used to with Fluent nHibernate.
Ultimately, if I start a business and I'm getting a lot of steam behind a certain project or concept, I would probably port it back into WCF with nHibernate if Silverlight was still requirement or go right back to WPF and nHibernate if not. The great thing about the exposure to Silverlight, is that Silverlight is a neutered version of WPF, so going in that direction should be easier than the other way around.
Anyway, thinking about a potential business is still a ways off. After DevTracker... the next project it will track is the Silverlight version of Veronica's software, provided my reporting solution will work in Silverlight. I spent enough time digging up that solution that I WILL scrap Silverlight for that side of things if I can't make it work. Then after that will be a modular but basic business accounting package, proper tracking of ledgers is something Veronica, myself and probably many of my currently fictional future customers would be interested in. And then after that... I have 1 or 2 projects I've always been tempted to start for fun. Both could make money, so they may be the platform for my future enterprises.
Anyway, tired or not, getting late and I need some sort of sleep if I plan on working 8 hours tomorrow.
Comments
Post a Comment