SLOCs and Other Statistics

The Reuters' project I’ve previously written about is finally coming to an end, at least version 1.0. It is now “feature complete” and tomorrow it will be deployed in their own servers. That’s a nice milestone!

Then I asked myself, how many lines of code have I written during this month? How many did I write per day in average? This is an unique opportunity to know, since I have been responsible for this little project from beginning to end, from analysis to design and implementation and testing, through the installation in their own infrastructure.

Just a quick reminder of the technical characteristics of the project:

One important detail: this application has been completely written “by hand”, from scratch; I have not used Visual Interdev to create it (I just don’t want to use it) and it exploits VBScript and JavaScript object-oriented capabilities at its maximum… which seems like saying hard work for little benefit right? Well actually the idea behind this design principle is to give Reuters an application that would be easily portable to .NET in the future. “Classic” ASP is a dead technology that will no longer be supported by Microsoft in the future…

The different modules have been separated in layers following a classic 3-tier architecture, and I used the following open source utilities to help me build it:

Here’s some statistics showing different checkpoints during the month of September; these have been done with Campwood Software SourceMonitor Version 2.0 (nice utility, quite awkward, but hey, it’s free :)

Date Lines (SLOC) # of files Cumulated working days Average lines per day
5 Sep 1075 20 2 537.5
12 Sep 1437 25 5 287.4
15 Sep 2745 31 8 343.1
19 Sep 4379 38 10 437.9
22 Sep 5245 45 13 403.5
25 Sep 6070 56 15 404.7
26 Sep 6235 58 16 389.7

Some observations:

OK, hope that someone finds this information interesting… I do! It was real fun to do this application. I would love to show you more about it but, you know, this is propietary stuff and NDAs apply here :)