Forums : Posts

Hello, Since ohloh counts code lines etc. in shell scripts, Makefiles, while are not exactly shell scripts, still have some shell code elements, so maybe they can be counted as well? I don't think a ... [More] separate category for makefiles is absolutely necessary, and they could be bunched up with the stats of shell scripts instead. What do you think? (Obviously, nothing urgent or priority here.) -s [Less]

OK, I've scheduled a new download. A new report should be ready later today. Let me know if it seems incorrect. Thanks, Robin

My question is how often (if at all) the repo is consulted to re-enlist the data from there because the repo changes over time. I think that was answered here: ... [More] (http://ohloh.org/forums/8/topics/85). Sorry about that. Could you please reschedule MARF: (http://ohloh.org/projects/3508) when you are not too busy. Thanks! -s [Less]

Hi moodler, This project report was indeed corrupted. We had successfully downloaded the complete CVS repository, but only a portion of it was inserted into our database. I've rooted out the problem ... [More] and prepared a new analysis. The new report seems to be what you expected. Thanks for the catch, Robin [Less]

Originally, the figures showed all of our contrib code as well, which distorted the figures a bit, so I edited the repository list to narrow it down to what people would normally call Moodle. ... [More] However, now the figures are really inaccurate. It says the Moodle project is only a year old, when the CVS repository goes back about 4 years. Hopefully it will straighten out, but in the meantime, ignore all these numbers. :-) [Less]

Hi Microcline, Great question. I hope to write an article on this very topic soon. The short answer is that Ohloh was created to provide transparency into the software development process behind Open ... [More] Source Software. This grew out of a consistent need I found myself having when I had to evaluate which piece of software to use/integrate into my own applications. I often found myself with basic questions like: How well maintained is this project? How many developers are contributing, how often, how much, etc.... How big of a project is this? Some smaller libraries are sometimes more of a pain to integrate than they are to just write yourself. Which of these competing projects should I choose? By looking at the amount of activity in competing projects, one can sometimes see that a smaller project has a much stronger curve/trajectory than established or larger projects. What languages is this written in? Using Open Source Software today usually means getting your hands dirty to some extent. When choosing a library/application to integrate, it's pretty critical to ensure that you'll be able to at least understand some basics behind it before moving forward. I don't know python that well so I tend to choose alternatives in other languages sometimes. What's the licensing status of a project? When you integrate OSS into your app, you assume some legal liabilities associated with that software. Even if the software claims to be MIT or BSD license, some codebases contain snippets of other licenses -- which can be a red flag. There's some more questions that Ohloh helps answer, but at this point I figure I'll save em for the article ;-). Hope this helps, -jay [Less]

Hi Mark, I noticed you have .adb files and .ada in that module. Our parser only recognizes .ada files for now, although this should be easy to fix. I'll take a look tomorrow to see why it's not resulting in any ada code.

Hi CyberBob2, Welcome to ohloh! We're essentially a wiki: anyone can edit. I went ahead and changed the enlistments to the svn enlistments shareaza now uses. The download is at 5%... looks like it ... [More] will take a few hours. Thanks for the notice. -jay [Less]

My apologies - the faq steps weren't specific enough. I updated them now. The key was to use siteexplorer's inlinks functionality (not the site pages -- as you were seeing). Okay... Another point ... [More] I will add to the FAQ is that the numbers seen from siteexplorer and from what we get using the Yahoo Search API are sometimes a slightly different. In MARF's case, for example, siteexplorer says 38 while the Yahoo Search API says 83. That's 2x difference! Please keep using the Search API :-D Thanks. -s [Less]

On the developer row we are listing the total of new lines added, without subtracting lines later removed. This means that if you add 5 lines of code and someone else later removes 3 of those lines ... [More] , you will still be credited with 5 lines added. So what's happening in this project is that neville is editing a lot of HTML and is getting credit for adding a lot of lines of HTML, even though most of those lines are later removed or edited. In absolute terms, this project is 60% Java, but along the way there has been a lot more editing of HTML than of Java. Why do we do this? Honestly, because it's computationally simple. When someone removes a line of code, it's very difficult for us to figure out which developer originally wrote that line of code so that we can debit the removed line from the old developer. To keep things easy, we just keep counting up added lines. We're not satisfied with this, and we're thinking of ways to improve things. Robin [Less]