## ossmole-discuss — discussion of ossmole general ideas, concerns, whatever.

You can subscribe to this list here.

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Jan Feb Mar Apr May Jun Jul Aug Sep (8) Oct (101) Nov (18) Dec (5) Jan (14) Feb (20) Mar (26) Apr (19) May (15) Jun (9) Jul (20) Aug (51) Sep (35) Oct (7) Nov (1) Dec (2) Jan (6) Feb (38) Mar (9) Apr (1) May Jun (5) Jul (4) Aug (8) Sep Oct (22) Nov (20) Dec (2) Jan (2) Feb (3) Mar (1) Apr (6) May (19) Jun (7) Jul (21) Aug (4) Sep (10) Oct Nov (2) Dec (8) Jan (10) Feb (5) Mar (1) Apr (12) May (7) Jun (7) Jul (15) Aug (22) Sep (15) Oct (26) Nov (41) Dec (13) Jan (5) Feb (11) Mar (8) Apr (15) May (14) Jun (43) Jul (10) Aug (2) Sep (16) Oct (33) Nov (16) Dec (7) Jan (3) Feb (27) Mar (15) Apr (16) May (7) Jun (11) Jul (8) Aug (6) Sep (11) Oct (1) Nov (17) Dec (7) Jan (2) Feb (6) Mar (1) Apr (9) May (14) Jun (3) Jul (4) Aug (1) Sep (3) Oct (4) Nov (15) Dec (5) Jan (1) Feb (5) Mar (10) Apr (4) May (2) Jun (10) Jul (34) Aug (9) Sep (2) Oct Nov (4) Dec (12) Jan (4) Feb (3) Mar (3) Apr (3) May (5) Jun Jul (1) Aug Sep (7) Oct (5) Nov (1) Dec (1) Jan (3) Feb (8) Mar (9) Apr (3) May (4) Jun (4) Jul (5) Aug Sep Oct Nov (2) Dec (2) Jan (1) Feb (2) Mar (1) Apr (2) May (2) Jun (3) Jul (3) Aug (2) Sep Oct Nov Dec (1) Jan (4) Feb Mar Apr May Jun Jul (2) Aug (2) Sep (1) Oct Nov (2) Dec (1) Jan (4) Feb (1) Mar (1) Apr May Jun (2) Jul Aug Sep Oct Nov Dec
S M T W T F S
1

2

3

4
(1)
5
(1)
6

7

8

9
(1)
10

11

12

13

14
(1)
15

16

17

18

19

20

21

22

23

24

25

26
(3)
27

28

29

30

Showing 7 results of 7

 Re: [Ossmole-discuss] SourceForge Activity Question From: Dirk Riehle - 2008-06-26 16:35:36 ```>> If the formula was correct, the max value would be: >> >> (1 + 1 + 1) / 3 + (1 + 1 + 1) / 3 + (1 + 1 + 1) / 3 = 300% >> > > I think that the (much expanded) forumla on that page calculates a > single figure, not a percentage. I guess they then order those > figures and calculate the percentile into which that figure falls, > which is where the 100% maximum comes from (a percentile)? > > Sorry if that makes no sense, I just glanced quickly. An example with > actual values would be useful. Of course being a percentile one > wouldn't be able to replicate the figure without access to all the > rankings. > No, that sounds like a reasonable explanation, i.e. they just don't mention the mapping from the range 0..3 to percentiles. Dirk -- Into novel software paradigms, tools, processes? Then submit a short paper to Onward! 2008 by July 2nd! See http://www.oopsla.org/oopsla2008/cfp/cfp-onward.html -- Phone: + 1 (650) 215 3459, Web: http://www.riehle.org ```
 Re: [Ossmole-discuss] SourceForge Activity Question From: James Howison - 2008-06-26 15:31:34 ```On Jun 25, 2008, at 11:33 PM, Dirk Riehle wrote: > Hello everyone, > > is this a good place to discuss SF metrics questions? Certainly :) > We don't quite understand the project ranking index definition at: > > http://alexandria.wiki.sourceforge.net/Statistics > > Each of the components (traffic, development, communication) can > reach a > maximum value of 100%. Then they are added to derive the final ranking > value, making it 300%. The observed values, e.g. Azureus, are 100% > max. > It looks more like they take the max. value of these three components? > > If the formula was correct, the max value would be: > > (1 + 1 + 1) / 3 + (1 + 1 + 1) / 3 + (1 + 1 + 1) / 3 = 300% I think that the (much expanded) forumla on that page calculates a single figure, not a percentage. I guess they then order those figures and calculate the percentile into which that figure falls, which is where the 100% maximum comes from (a percentile)? Sorry if that makes no sense, I just glanced quickly. An example with actual values would be useful. Of course being a percentile one wouldn't be able to replicate the figure without access to all the rankings. > Also, are project rankings equivalent to activity ranking? Like > Daffara > uses it here: > > http://robertogaloppini.net/2007/08/23/estimating-the-number-of-active-and-stable-floss-projects/ I think someone from the FLOSSmetrics project would be best placed to answer that. I don't think that Roberto is on the list. Jesus or Gregorio, could you invite Roberto? (btw, do you guys provide a workflow for this classification?) --J ```
 [Ossmole-discuss] SourceForge Activity Question From: Dirk Riehle - 2008-06-26 03:33:00 ```Hello everyone, is this a good place to discuss SF metrics questions? We don't quite understand the project ranking index definition at: http://alexandria.wiki.sourceforge.net/Statistics Each of the components (traffic, development, communication) can reach a maximum value of 100%. Then they are added to derive the final ranking value, making it 300%. The observed values, e.g. Azureus, are 100% max. It looks more like they take the max. value of these three components? If the formula was correct, the max value would be: (1 + 1 + 1) / 3 + (1 + 1 + 1) / 3 + (1 + 1 + 1) / 3 = 300% Also, are project rankings equivalent to activity ranking? Like Daffara uses it here: http://robertogaloppini.net/2007/08/23/estimating-the-number-of-active-and-stable-floss-projects/ Thanks, Dirk -- Into novel software paradigms, tools, processes? Then submit a short paper to Onward! 2008 by July 2nd! See http://www.oopsla.org/oopsla2008/cfp/cfp-onward.html -- Phone: + 1 (650) 215 3459, Web: http://www.riehle.org ```
 [Ossmole-discuss] google code releases From: Megan Squire Conklin - 2008-06-14 16:32:31 ```Hi moles, I am in the process of moving our releases to Google Code from Sourceforge. Ran into a bit of a snag this month with the datamarts as they have a max file size limit of 20 MB. Our SF files are about 3 times that big. Here are the downloads so far: http://code.google.com/p/flossmole/downloads/list I've submitted a ticket to get the quota increased. http://code.google.com/p/support/issues/detail?id=1139 So far it is much, much, much easier to release code to Google than it ever was to release to SF. There are a couple things missing at Google which I did like, such as release notes, and so forth. But so far, I like it a lot better. (python release from commandline is sweet, for example) I'll keep you posted on the move, and whether other things, like wiki and mailing list, will move or if we'll keep those at SF. -megan ```
 Re: [Ossmole-discuss] Beginnings of analysis of Codeplex (viz Sourceforge) From: Megan Squire Conklin - 2008-06-09 14:36:28 ```Thanks James, that is very interesting. I have been in conversations with Microsoft regarding Codeplex for a while now, including with Vlad and Marc, but they have stalled for an unknown reason. Unfortunately, our last con call was supposed to be on April 22 but no one showed up except me, ha ha :) So I'm trying to get the group back on track - I have pinged them just now - and hopefully we can get back on track with, at the very least, the Codeplex data dumped into FLOSSmole on a regular basis (simliar to the XML Freshmeat dumps, was my hope). FYI, the people on the con call list at Codeplex were: Marc Smith; Vladimir Barash; Mario Madden; Sara Ford; Jonathan Wanagel Fingers crossed! This seemed like a good data set to have and some interesting people to work with. -megan James Howison wrote: > Here's an email I get from Marc Smith from Microsoft Research that is > always interesting, but this week he discusses the start of an analysis > of Codeplex, with Vladimir Barash, and the start of comparisons with > Sourceforge. > > Apologies if the formatting is a little screwed up. > > --J > > Begin forwarded message: > >> From: Marc Smith >> Date: June 4, 2008 9:29:39 AM EDT >> To: Marc Smith Weekly Report >> Subject: Marc Smith Weekly Report - for the 30th of May 2008 ```
 [Ossmole-discuss] evidence that devs are interested in similar questions to researchers? From: James Howison - 2008-06-05 00:12:58 ```http://kerneltrap.org/Linux/One-Time_Contributers --J ```
 [Ossmole-discuss] Beginnings of analysis of Codeplex (viz Sourceforge) From: James Howison - 2008-06-04 16:03:39 Attachments: image004.jpg text/plain image006.jpg text/plain image008.jpg text/plain image009.jpg text/plain image014.jpg text/plain image016.jpg text/plain image021.jpg text/plain ```Here's an email I get from Marc Smith from Microsoft Research that is always interesting, but this week he discusses the start of an analysis of Codeplex, with Vladimir Barash, and the start of comparisons with Sourceforge. Apologies if the formatting is a little screwed up. --J Begin forwarded message: > From: Marc Smith > Date: June 4, 2008 9:29:39 AM EDT > To: Marc Smith Weekly Report > Subject: Marc Smith Weekly Report - for the 30th of May 2008 > > > [Marc Smith: Week in Review] > http://autogroup/JoinGroup.asp?GroupAlias=mswkrpt > > [Recently] > > > [Vladimir Barash - SVC internship week 1!] > > ```

Showing 7 results of 7