bskool bloggin

Making More Sense

While XML has proven to be a complex and somewhat theoretical experience over the last week and a half, I had somewhat of an epiphany today when we ran query 1 and multiple rows were in the resulting row set. I said to myself 'self, wala!, that looks like HTML almost with all of those lines with little tags and varying indents.' This made me realize what Bud meant by saying that XML can be used to take data and turn in into a lot of useful things. We have very raw data on the Oracle server which we can easily (relatively, at least) now query and display in useful ways like HTML pages. Wow. Sorry if everyone else realized this when we talked in class the first day, but today was the click. I couldn't write it all the code or define every term, but the concept and use is now much clearer.

Adding onto the XML Conversation

I found Jrtrana\'s blog post about XML to be inspiration to check out a bit of XML.com myself, but I found it to be quite confusing. I never made myself learn HTML, but I do understand that HTML has various tags which define what a particular part of the page is. The idea of having no defined tags seems strange and I will be interested to see how you define them and if that is a lot more work than having defined tags already in place.

The author of the "What is XML" article on that website also points out that the documents that XML is used for would simply not be appropriate for HTML. Again, I'll be interested to see what types of documents we're talking about. Should be an interesting topic, but I find myself currently quite in the dark.

ER 2 ?

In ER Excercise #2 the answer to question #1 states that advisor, which is the 1-M relationship between professor and student needs to be represented by a table, while position, which is the associative entity between professor and department does not need to be represented by a table. This seems contrary to our rule that M-M relationships, entity types, and associative entities are the only things we respresent as tables. Also, in the question #2, the answers say that a foreign key of professor needs to be embedded in the advisor table.

SQL'd OUT!

Well it may just be me, but the SQL excercises and chapters seem to go on forever in the new book compared to the coursepack readings and excercises on ER Diagrams and Normalization. I don't know if it is because there are just a lot of new terms and topics in a shorter amount of time, or the month of October's added stress is just making it seem worse.

As a note for anyone using Professor Gibson's scripts to check your answers, there is some different terms that he uses which you have to be careful of. For example, many times he uses "o.order_number" instead of "order_num" like I had it loaded into my table. Thus I was getting a correct answer with the script I wrote, but his script wasn't running for me.

Also, I noticed that there is often more than one way to write these queries in the way they want in the book, but that the "answer" can look different. I believe that the exam will allow for different ways of writing queries that get the same result, but I just realized that everyone probably picks up their own style eventually. For example, I felt that using aliases often required more time and energy that they were worth for short queries, but then for longer queries, with more than two tables, they were a real help.

Never Mind Me

And as with most stupid questions I think I've answered myself. I am going to follow what they are doing in the book exactly on my own by creating 'ex1,' with is creating table rep and then dropping that table using the "@" in Oracle.

On a Guide to Ch 2

Can someone explain to me what we are supposed to do for the pg 27-29 bullet point on the sheet "Notes for 'A Guide to SQL' Chapter 2" where we are supposed to create a command file in jedit? What do you put in the file? "Create Table?" When I do that and then run @exl in the ssh window it prints out a 2. Is the command supposed to be in the file or is the comand the file itself? Sorry if this is entirely confusing, but I'm not quite sure what we're doing here as my database experience has been completely in Access.

No Fun Feedreaders

By the way, it seems as though I will no longer have fun seeing people's new formating on their blogs due to the use of my feedreader. Plain text gets boring....

The question is whether feedreaders work for blogs that incorporate a lot of media-rich content like pictures. Do you have to go to the blogs themselves if people are incorporating this stuff?

On Bud's Blog

With regards to track backs do I need to post on my blog and put in a web link to the other person's blog for them to get a trackback or do I post a comment on theirs and then put in my URL where they have a comment section? This whole idea is confusing. I posted on Bud's Blogonaut site under his CBS story. Here is the link

Regarding Printing

Since this little part of the blogosphere is all business school students I thought I'd bring up the issue of printing. I'm glad that we do not have to print anything out for BIT320, because in the rest of my classes I'm going crazy trying to print in a hurry with the new Xerox machine system. Standing in line to swipe my card is a serious waste of time, and two printers when there used to be four in the Kresge lab makes no sense at all. Hopefully they will find creative solutions to these issues early in the semester, as it could become quite an issue during final projects and exams when students need to spend less time printing, and more time on value-added to their studies.

Regarding My Question

In case my question in class regarding normalization sounded ambiguous, I wanted to clarify. I asked "do we peform normalization only when we have a really messed up database? is this a service offered by some firms?"

In speaking with Professor Gibson after class and reflection on the topic, I basically realize that this is a very useful concept with the everyday data issues that analysts have to deal with, at least in consulting projects like the one I worked on over the summer. In my case I had to work with a data tool that was made prior to me coming on board. There was a lot of repeat information, and some fields were like the 'child x' example (for instance primary contact and secondary contact, where there wasn't necessarily a #1 and #1 person for the suppliers that I was modifying in the database). With the concepts we've discussed I would not be able to speak with one of my company's data people to better display information so that my use of it would not be stifled with confusion.