Design and Architecture

This would really depend on how "in the weeds" your developers are getting. As a developer I find myself installing and removing applications almost every other day for integration testing and evaluating new development tools. In addition, I have occasion to stop and start services on my development box when things go haywire. Editing and removing registry entries is not uncommon as well. All of these are much easier with local admin rights. If your developers are doing simple web development (HTML, Javascript, Flash..etc) then they probably don't need admin access. However, any of the higher languages (Java, C/C++, C#, Perl...etc) I would definitely recommend admin access. One solution I have seen to your dilemma is install virtual environments. This allows the developers to create their own machines with what ever rights they want, while still controlling the actual hardware. Unfortunately, the virtual env. tends to be slower and can affect compile and testing times. My guess is that your virus scares were not on the developer boxes, they tend to be more careful than your average user. This fight is not uncommon. modified on Thursday, January 7, 2010 10:27 AM

Create run time stage table depending upon the format of targetting file.....

Need to define the scope of the problem... which part you are looking for....

There is no need of reinstalling of window service there...... But Question regarding business components... depending upon the type of assembly... if it is deployed in GAC... there is not need to recompile the window service... else need to recompile the service....

LINQ 2 SQL, AWESOME for MS SQL small to medium size projects. Easy to use and implement. Entity Framework - Pain in the arse, 100%. Lightspeed - Commercial solution but VERY awesome. More high-powered than LINQ 2 SQL, with more flexibility (+ compatible w/ other DB engines). NHibernate - Too much XML hand editing, makes it slow to use and to modify sometimes.

I think I will go with this approach. Like you said, it gives less room for error ("pit of success" and all that).

Please do reply back of what you think about the approach or is there any better approach to cater to my requirement.

you have no friggin’ idea what this frigtard is talking about, but he just won’t go away Well said by Joel. I've worked with these types of individuals in the past and can relate. "The clue train passed his station without stopping." - John Simmons / outlaw programmer "Real programmers just throw a bunch of 1s and 0s at the computer to see what sticks" - Pete O'Hanlon "Not only do you continue to babble nonsense, you can't even correctly remember the nonsense you babbled just minutes ago." - Rob Graham

Apart from not knowing exactly how a lower layer does something, "removing dependency" could also be considered as "not knowing WHO is doing what I need to do", and that is acomplished using Interfaces. So you go from not only not knowing how, but neither knowing which class is the one that implemented the interface and is executing whatever you need from the lower layer. Why would you do something like this for a DAL? Two main reasons, first not to depend 100% on an implementation for a specific database engine (ie SQL Server), but to have an interface that could be implemented for different storage mechanisms; and second for testing purpose! Imagine you could automate your tests by implementing a piece of your DAL (interface) from a dummy class (either mock or stub) just to test your upper layers (probably business).

Hello to all, it's me again. Reading more and more, I begin to doubt my own statements in my last contribution. Here I wrote under point b): " ... no arichitectural elements .... It (SRS) deals only and exclusively with the problem-domain. The goal is to structure the problem". Is this really true? Does a SRS really have no SW-architectural Elements? Is it's only purpose to structure/understand the problem-domain? If this is true, then: Where are the models for architecture coming up? Best regards

the CP rules state you choose one appropriate forum and ask your question there; don't post the same stuff in multiple forums, it will not help you. :) Luc Pattyn Have a look at my entry for the lean-and-mean competition; please provide comments, feedback, discussion, and don’t forget to vote for it! Thank you. Local announcement (Antwerp region): Lange Wapper? Neen!

Thnx buddy :)

I would post this in the C# section if I was you. It is not a design and architecture question so you are unlikely to get an answer here.

jarajeshwaran wrote: The application will be very simple with not more than 20screens. The underlying database will contain not more that 25tables. That doesn't add up; I'd expect a "list"-form and a "details"-form for every entity, allowing you to browse the entire collection or edit a single entity. jarajeshwaran wrote: Even if it is a multiuser application the number of user will not be more than 5 at a time. The application will be used by a large number of people so bug fixing and introducing new features should be a breeze New features could be introduced with a plugin-system, you could opt to load an assembly dynamic. There are some articles on CP explaining how to do such a thing :) jarajeshwaran wrote: The app will be having basic CRUD screens and the Business logic involved will not be too complex. So a generic CRUD-form would suffice? jarajeshwaran wrote: The time to market of the resultant should be very less. I'd walk away as soon as this sentence drops, without any hesitation. The TTM should [b]never[/b] be confined by the manager. Imagine you selling a house with a limited TTM: would you sell it without a roof? --edit-- As I've been told, the houses are sometimes already sold before the drawings are even agreed upon. There seem to be special constructions to ensure that you get a roof. And that construction won't grind to a halt midterm. jarajeshwaran wrote: I would like to know which framework or design pattern should I use to achieve the above functionality with least coding. I don't want to go with any third party tools. The .NET framework, and databinding :) I are Troll :) modified on Friday, September 18, 2009 4:09 PM

I have a DAL and then a single entity object, problem seems to be that a collection of the entity objects has one hell a lot of baggage in it. I'm currently working through Josh Smith's[^] article on MVVM and wondering if the level of complexity is valid for LOB apps. I look at it from a down stream support aspect. I am slowly coming around to the opinion that WPF/Silverlight is probably going to be the future UI to work with and the MVVM pattern seems to have a vote of confidence from some well respected people.

Maybe some time studying UML would be the best place to start.