Dear Open Hub Users,
We’re excited to announce that we will be moving the Open Hub Forum to
https://community.blackduck.com/s/black-duck-open-hub.
Beginning immediately, users can head over,
register,
get technical help and discuss issue pertinent to the Open Hub. Registered users can also subscribe to Open Hub announcements here.
On May 1, 2020, we will be freezing https://www.openhub.net/forums and users will not be able to create new discussions. If you have any questions and concerns, please email us at
info@openhub.net
First off, wanted to say that I'm delighted to see the new username aliasing feature. Nice work, great interface, good stuff. That was probably our #3 ohloh gripe, with #1 being listed below and #2 sounds like is under development (ignoring dirs). :-)
On to the main point, I was wondering if there has been any additional thoughts or considerations put in to having Ohloh process all RCS revisions on HEAD instead of just the latest. This is a follow-up to the discussion held a month ago over here. Basically, it boiled down to using:
cvs log -b -r1:
Especially significant for older projects, this captures HEAD activity comprehensively and shouldn't affect statistics for the majority of codes that have never changed their revision number. It would, however, resolve our particular activity discrepancy and contribution statistics.
Thanks again for fighting the good fight, implementing good code and for providing a great overview interface with useful statistics.
Cheers!
Sean
BRL-CAD
Hi sean,
I am in agreement with you on this change, and I have not forgotten you -- quite the contrary.
A change to our CVS downloader will ultimately require re-downloading all of our existing CVS projects and rebuilding our database basically from scratch -- a huge impact to our operations. If we're going to take that leap, we want to batch together many other fixes we're making to get the most bang for our buck.
I know it's dissatisfying to have outstanding issues, but right now there are a lot of changes happening behind the scenes to keep our service alive and moving forward. Looking at the work on our plate right now, I'd guess that it will probably be several weeks before we have opportunity to crack open the CVS downloader code for revisions.
I really appreciate your detailed help, kind words, and patience. I'm typing as fast as I can!
Robin
Hi Robin!
Thanks for the info and the reply. It's great to hear that the feature/change will eventually happen and understandable how you need to plan for it. Sounds like I just need to get out of your way now and stop slowing you down. :-)
Cheers!
Sean
Hi sean,
It's been a long time, but I have finally found some time to work on updating our CVS functionality as described below.
I'm currently running a complete, new download of BRL-CAD with the '-b -r1:' options as shown below. The download will probably take a few days to complete, but I wanted to give you a heads-up about the download in case something peculiar happens to the BRL-CAD report along the way.
I'll be keeping a close eye on this job over the next few days. If all goes well, the '-b -r1:' option will be rolled out to more projects.
Let me know if you see anything unusual. I'd appreciate your help in verifying the report when it's complete.
Thanks,
Robin
Howdy Robin!
Actually, I noticed about a week ago right after you'd started that all the stats were updated! I seem to find myself visiting the site for various statistics more and more frequently. Needless to say, I was quite delighted. The stats for BRL-CAD look much better now.. THANK YOU so much for all your hard work on running the update.
Comparing the final ohloh user stats that you pulled with the stats that I extracted, the numbers were very close.. They weren't matching, though, as I would have expected for at least the older devs that are no longer active so I dug deeper. Fortunately when I did, though, I was able to account for all of the differences and got exactly the same counts for all the devs I compared. The difference was that my stats collapsed all identical consecutive log messages contrary to ohloh's (better) time-based collapse. Once I took that into account, the numbers matched up.
From a larger sanity check standpoint, I counted 25836 unique commits where ohloh has 26554 identified, which is pretty much what I would expect given the time-based culling differences. Another sanity check was to look at the year's contribution of the developers (which started this whole inquiry), which also seems much more reasonable now.
Thank you again for putting the effort into this and getting the update rolled in. If there is a donate now
button or other means to convey my gratitude and appreciation, please let me know.
Cheers!
Sean