Other Articles‎ > ‎

Help Simplify Life

It was in 1990 give or take a year that I read about a company which had created its entire application using a spreadsheet. The spreadsheet was Wingz and the platform Mac. Unfortunately, I could find very little about Wingz now. It was a product from Informix with a “powerful” scripting language as its USP.

This story re-emerges frequently in my mind especially when we work with tabular data, which is almost all the time. The difficulty of getting the reports formatted correctly has been a source of contention with the user departments far too often. As an experiment, we had created macros by which an IT literate HR executive, Vishnu, could selectively load data from the database into Lotus123 and process it the way it was needed. It worked very well. Vishnu could finish his job faster than it would have taken him to explain his needs to the programming staff. His productivity and effectiveness increased and he could respond quickly to management needs even when confidential processing was involved. Needless to say, Vishnu was very happy and enthusiastic about the project.

OpenOffice has componentized architecture and can be programmed using UNO. PyUNO makes it possible for us to program and control OpenOffice across the network using Python. Hence, my plan was to get started with PyUNO and, I expected, that like everything else in Python, each step leads to the next and pretty soon I will be able to advise people about how to use OpenOffice as the 'user interface' and use Python for programming the logic.

It hasn't worked out that way because I have failed to take the first steps in the free time slots I have been able to manage. Meanwhile, in my search, I came across this link, along with a critical extract, on Ubuntu's web site:

http://www.ubuntulinux.org/community/bounties

The 2004 budget for bounties is USD 100,000.

Bounties will be offered on Python scripting interfaces for the following tools:

  • OpenOffice

  • Blender

  • AbiWord

  • Gnumeric

  • The GIMP

So, there are other people wanting the same thing as I want and are willing to pay for it as well! Hopefully, the budget for 2005 will be still larger.

I will have to learn how to use OpenOffice components using Python because I plan to use it as an example in my class. Meanwhile, I would like to spread the word and encourage others to contribute to PyUNO and make it simpler to get started.

For example, OpenOffice comes with its own Python interpreter which is version 2.2. There is a need to bring the level to the current. It would also be worth looking into and finding whether OpenOffice should use the default Python installation, at least, if Python is installed. If nothing else, it will create a slightly slimmer download of OpenOffice.

Why should one spend this effort?

I am sure there are many who have been irritated by having generate variations of the same reports with a few columns present or missing or formatted differently, all which could be trivially handled by a spreadsheet.

Or, imagine using a program to create a skeleton presentation of the performance of various offices. The presentation contains all the slides with the actual performance figures presented just the way the management wants. It can then be extended by each office with supplementary slides explaining the significance of the results. The managers should focus on adding value to the interpretation of results rather than spending a ridiculous amount of effort on purely presentation issues (frequently with the explicit intention of obscuring the results).

Bureaucracy is everywhere. It wants, no, demands, a report - typically a document consisting of data picked from a dozen locations organised in a very specific manner. This, again, is a step requiring enormous effort and energies just to collate the various data. The process is heavily manpower intensive and, far too often, inconsistent, obsolete or incorrect information slips through.

In each of the above cases, it will, in general, save us time if we could program the extraction of the dynamic data from various applications and its insertion into an OpenOffice document. Once we know how to do it and the appropriate tools are available, I am sure it will be no harder than creating a dynamic web page.

I hope I have been able to incite just a few of you with the scenarios above. I can assure you that thousands, may be millions, of people will be eternally grateful to you for saving them from the drudgery and agony of tedious, monotonous activities.



Comments