funny you mention lisp... just the other day i wrote a AI app (which is purpose of lisp) in LISP.

Arsen

Hello WT, The situation you are describing is a very common situation in development. We all go through managing the changes to code and consolidation of these changes. Here is a suggestion if you are referring to source control for developers: For application code (C#), use a source control system. In my organization, I set up to use Microsoft Team Foundation Server. Basically, its pretty cool in that when developers make changes in the C# code, then the changes are automatically checked out and then they can check things in. Other source control systems you can use can be Visual SourceSafe. Or you can go with a open source solution such as: SubVersion. Now for SQL scripts(I assume that is what you mean). There are a few solutions out there which are add-ons to SQL 2005 to handle source control. As far as deploying new SQL changes (which seems to be the core problem) to futher environments. You can use a SQL examiner tool which would do a compare between DB1 and DB2 and then allow you to create a "change script". I use SQL Examiner: <a href="http://www.sqlaccessories.com/SQL\_Examiner/"> </a>[<a href="http://www.sqlaccessories.com/SQL\_Examiner/" target="_blank" title="New Window">^</a>] I hope above is helpful for you.

Arsen

Ah.... software architecture. Very good question. There really isn't a clear definition, although there is a informal standard (yeah, these 2 words don't go well together). Software architecture can be defined as a design and blueprint of how the software components will work with each other, which objects are part of the software, how objects will interact with each other, and so on. There can be a high-level architecture which may include: Tier layers such as Database, application and UI. And there can be a low-level architecture which may include: Objects (classes) which are part of the software solution. Most good software architects design the entity relationships and classes after designing a good and clear functional design. A process in which I follow: 1) Design clear functional specifications which include detailed use cases of how each requirement will be implemented. (actually this is more of a job for a good technical analyst) 2) Design the component, state, process and class diagrams as detailed as possible. Basically the end result should be a clear blueprint of how the software should be developed. Therefore a developer would just do the "implementation" and the design to have already been done by the software architect, hence the developer becomes a "robot" following the instructions of the architect. I hope above helps.

Arsen

SOA is a interesting concept, yet it "technically" isn't anything new. Its really another type of distributed computing architecture environment. Conceptually SOA was around since the 80's, but just recently (in past 4-5 years) the tools and technologies have become available to really use these concepts. There are 2 types of SOAs: True SOA and False SOA Most people think that SOA is just a set of web services. That is "false SOA". It doesn't mean that is not SOA, it is SOA, but not utilizing all architecture concepts. Designing a service-oriented-architecture requires designing the architecture from inception. During the design phase, one determines which components and features can become a web service. It is best to try to make as many features of the system a web service. Basically, creating another type of interface to the business logic. Here are a few good books on this subject: [^] In summary: SOA Is: * An approach to information interoperability across diverse systems. * Architecture style/concept * Transformation to a business & IT * The next Generation Architecture will constitute the third big era in the It industry's history. * SOA uses interactive business components designed to be meanfulful, usable and useful across application or enterprise boundaries (integration + interoperability) SOA is NOT: * New * A technology * A Web service * A specification * A set of standards * Code * Just a hype or a buzz word Good luck!

Arsen

Here is what my developer has implemented (excerpts) void ProcesstopicChildren(DataTable dttopic) { similar while loop, but calling processtopic within whileloop. } main() { DataTable dtChildtopic; DataTable dtParenttopic; ItopicManager topicManager; // this is the SQL connector to call the SQL. int ParenttopicID; ParenttopicID = 8; // Just as a test dtParenttopic = topicManager.GettopicView(ParenttopicID); while(dttopic.Rows.Count > 0) { ParenttopicID = int.Parse(dtParenttopic.Rows[ParenttopicCount]["ParenttopicID"].ToString()); dttopic = topicManager.GettopicView(ParenttopicID); ParenttopicCount++; } }

No not homework at all... actually one of my developers is having trouble with this, he wrote the code but i'd like to get some opinions from others on what patterns they use.

Arsen

Greetings, I will try to describe this problem in as much details as I can. Task: Create N HTML page withing a folder-structure based on N grandparent-parent-child relationships. Data structure: TopicID - int ParentTopicID int TopicName - nvarchar(4000) Sample data TopicID ParentTopicID TopicName 0 0 Computers 1 1 Programming 2 1 C# 3 1 C++ 4 1 Java 5 1 Perl 6 0 Operating systems 7 6 Windows 2000 8 6 Windows XP 9 6 Unix Output which to achieve: Folder: Computer File: Computers.html Contents: Programming Operating Systems Folder: Computer/Programming File: Programming.html Contents: C# C++ Java Perl and so forth.... The question is, what is the best algorithm to use to achive the above results? Thank you very much in advanced!

Arsen