Deliverable 2 – INFO 320

Write-Up – Part 7

Ryan Prins

November 26, 2003

Revised Strategy

Bills of Lading & Plant Stock Information

For the second deliverable there had to be changes that were made to the structure of the XML as well as the way that the XSL style sheets were presented to the user. With the outcome determined by the project, flowers sorted into colored pages, some transformations were needed to create the desired output for the user.

The first of these transformations had to come within the XML files themselves. This deliverable provided us with 5 new XML files to work from. There was the bill of lading and the 4 individual XML files for the different flower types. These XML files were not handed to us in a format that we would be able to use directly. So, from a C# script we had to be able to pull the relevant data from the file and then create a new XML structure for our flower data. A very similar data structure was used from the last deliverable. The only changes were the removal of two elements (price & quantity) and the addition of a new attribute (color).

From deliverable one there was some excess data that was not needed. This data had to do with the pricing and quantities of the flowers. I choose to not provide this data to the user in the XML file since this data was not provided for the new flowers, so I was not going to have the data for some flowers, but not for others. I figure that it would be confusing to the user on why some had it and some didn’t. Plus, the quantity and price was not relevant information for the desired outcome, so it was completely left out for the sake of simplicity. But, if this data is needed in the future, a similar procedure of using a C# script to generate XML files would be used to get that information back to the user. This process would not be difficult, but it would still need to be done to add this information back into the XML source.

Because the new data came from many sources just made the addition to the one source a little trickier but not that extremely difficult. You needed to look at the bill of lading for some information then cross reference the specific flower XML to find the other information that you need before adding it to your final XML. I did not make “temporary” XML sources to store the data. I just took the original data, altered it as needed, and then added all of the new data to the bottom of the source. I did not see any reason on why this could not be done, or if there were any side effects on why it could not be done. In the end the transformation worked out as planned and the structure that I had planned on was maintained.

Personnel Data

As for the personnel information, I found no use for it on my site. There was no need to cross correlate the data with anything that was on the site. It would simply become excess data that the user would have no use for and by providing this information that serves no purpose at this time would not benefit the consumer of this product. So, it is best left out for this project, but addition to later project would require little work for proper implementation.

Since this personnel data is never used in the presentation, no new designs were needed for this information. In later projects this information may find a place along side the store information. This would be my first idea on where it should go since the personnel data would easily correlate to the store information data. However, this is just an idea and it could be easily changed at any time to accommodate for a specific output that is desired.

Sales Information

In this deliverable the only use of the sales information came when the Access database was generated. Again, as stated for the personnel data, this information served no purpose to be displayed on the site and therefore was not included for output in a XSL style sheet. This information remains idle in the original XML file and in the Access database, but if there was a further need to provide the data on the site, a simple modification could be made to the site to provide this needed change.

More than one XML source?

For this deliverable I have two XML sources. I have a XML source for the store information (hours, address, etc…) and for the flower data. This keeps the inventory separate from the store data and vice versa. This information was linked in the style sheet by the use of the document function. This provided access to the store information when it was needed on the flower pages, but at the same time kept the information separate from each other when looking at the raw data.

I prefer this method of multiple XML files because it keeps the data separated and ordered how I feel would best benefit the user. If I had to pass on my inventory to another user, they would care less about my store information, and maybe in some cases I might have information in the sources that I do not want them to see. So, I would have to make another transformation to get all of the flower data out of a single XML source. So, keeping these separate will allow for an easy transfer of information to another store or to another user who wishes to look at our inventory data.

The Future – Ad Hoc

For future data additions data would need to be looked at carefully on what would needed to be added where and if major modifications would be needed. If, for instance, the user demanded, and provided, data for the sale prices and quantities, I would need to add elements to my XML inventory source and funnel in this data with a C# script. This process would be similar to what was accomplished for this project, but would serve a different purpose than this insertion did. But, this could be done. It all depends on what the desired output is. But, the basic structure that I have setup provided easy access to insert data for any type of flower and it would be very simple to add an element that would provide extra information about the given flower. I spent some time thinking about the best way to set up the structure and I feel that my current structure is best for what has been asked of the XML source. It has been easy to get the data from it and to pass it along to other sources, if needed. I do not see this changing in the future, and if changes are needed, I anticipate that they would be minimal at best.

What about the re-use of information?

I have found in this project that the re-use of information is very important when providing a specific output to a user. It provides much greater flexibility for the user to change one file and to let that file filter the changes down to other files. This, in turn, creates a more rounded way of creating pages and providing the user an efficient and reliable way to change information across many pages. Also, with the simple use of attributes and variables, one can create truly generic pages that each are unique to each other. In our case, have one style sheet for the colored pages and pass variables into it to make it unique to each flower color. If using multiple style sheets, you will find that you need to change every change that you make to hard coded data in every file. This would become very tedious if there are many style sheets. So, by having a high reuse of information one can create a change and just re-export the file out again with the changes made to it. This will then make making changes much easier on the user when a change is passed down from the information.