issue backlog

118 views
Skip to first unread message

Philip Durbin

unread,
Oct 17, 2016, 9:38:53 AM10/17/16
to datave...@googlegroups.com
There are 704 open issues at https://github.com/IQSS/dataverse/issues and over the weekend I spent many hours looking at every single one of them (!) and categorizing them[1].

Next I will be taking the following actions:

- Agitating that certain issues be closed.
- Agitating that certain issues be consolidated.
- Flagging certain issues as "low-hanging-fruit"[2]
- Following up with users to see to get updates.
- Linking issues as appropriate.
- Making comments as appropriate.

Obviously, I have code to write, but I'll chip away at these while waiting for Glassfish to redeploy code I just wrote. :) [3]

Here are some observations about the corpus of open issues:

- Sadly, quite a number of issues have been opened by real users with real problems or suggestions that haven't even received a comment. Many of these bugs are spot on and many of the ideas are grand. I'll at least leave a comment on these thanks the user for their time!
- We should be using more granular "component" labels so that when part of the code is being worked on, other related issues could be knocked out as well.
- There are many "suggestion" issues that are really just placeholders for people to weigh in on an idea for a feature. These are somewhat hard to find.
- Developers like to open issues for their pet refactoring project that end users probably don't care about. :)

Ok, my Glassfish is up and running again. Gotta go! I'll leave you with two somewhat conflicting things to read about open issues:

- http://blog.atom.io/2016/04/19/managing-the-deluge-of-atom-issues.html
- "the higher the number of bugs in the database, the better the project looks" http://producingoss.com/en/getting-started.html#vc-and-bug-tracker-access

Happy Hacking!

Phil

1. I wonder if Consilience could help with this task.
2. For my definition of low-hanging-fruit, please see https://github.com/IQSS/dataverse/issues/2594#issuecomment-253815443

Philip Durbin

unread,
Oct 21, 2016, 10:26:02 AM10/21/16
to datave...@googlegroups.com
I've been asked for a list of the categories I put the issues into and which issue was put in what category. All the categories are in the Trello board I use to plan and organize my work: https://trello.com/b/uggFGv2H/philip-durbin

Here are my categories (you can ignore the other lists in that Trello board):

- Comments to make
- Vote to close
- Follow up with user or issue creator
- Refactor (backend stuff users don't care about)
- Retest (feature or bug I'm not very familiar with)
- Consolidate (consider combining related issues, possibly closing some)
- Link (don't necessarily consolidate)
- Low hanging fruit, perhaps, at least for me
- Legit bug or worthy project

I do still plan to make comments on various issues (including comments that say "I vote we close this issue because...") as time allows.

I hope this is helpful. If you have any questions please swing by my office or ping me at http://chat.dataverse.org

Thanks,

Phil

Philip Durbin

unread,
Jun 25, 2017, 10:10:09 PM6/25/17
to datave...@googlegroups.com
I took another more targeted pass through some of the open issues, especially the ones that were open by real users rather than us developers. :)

Since we were asked at the community meeting to create a "help wanted" label, I went ahead and made one and described it at https://groups.google.com/d/msg/dataverse-community/nMdEGUCOyzE/RtOG3a_HAQAJ

I think the "help wanted" label should be used with care. If you apply a "help wanted" label to an issue, you should be willing to be a mentor for that issue.

Phil

Philip Durbin

unread,
Jul 12, 2017, 7:06:06 PM7/12/17
to datave...@googlegroups.com
It did my heart good today to hear, "i'm liking these "persona" labels =) ...  because I can select a given persona and see issues with which I think I could help."

That was from http://irclog.iq.harvard.edu/dataverse/2017-07-12

Let me tell you about these new persona labels.

In my ongoing attempt to listen to what our users are telling us by opening GitHub issues, I started adding the following persona labels to open issues:

- Persona: Curator
- Persona: Depositor
- Persona: Sysadmin
- Persona: Superuser
- Persona: Downloader
- Persona: API User
- Persona: Hackathon Participant

I'd say these personas are still very much in flux, but if they're helping contributors find issues to help with, great! Mission accomplished!

If any of the personas don't make any sense to you or you thing we should modify them, please let me know!

If you click the IRC log link above you see me start talking about Silly Putty Day, a thought experiment that lead me to create these personas in the first place. What if once a year, we celebrated Silly Putty Day by picking an open issue to work on completely randomly.

First, we would print* all the open issues and tape them to the two-story glass wall in the Quantina. Here's what that wall looks like if you aren't familiar with it: http://i.imgur.com/IgtJSAf.jpg

Then, we roll some silly putty into a ball and chuck it at the wall, hitting one of the issues.

We pull the issue with the silly putty on it off the wall and ask ourselves, "What type of users will benefit from the resolution of this issue?" That's what the new persona label is supposed to answer. Resolving the issue will make curators happy, or depositors happy, or sysadmins happy, etc.

Wait a minute, you're saying. The numbers don't add up. We have 681 open issue and the persona labels have only been added to 500 of them. What gives? Well, probably some new issues have been added. Also, some issues really don't seem to apply to any of the personas above. A lot of them are things that only developers care about. So maybe we need "Persona: Core Developer" or something. I'm not even sure what we call ourselves so I figure we can add this persona later, if need be. To find all the issues that don't have any of the persona labels above, try the following query at https://github.com/search

repo:iqss/dataverse is:issue is:open -label:"Persona: Curator" -label:"Persona: Depositor" -label:"Persona: Sysadmin" -label:"Persona: Superuser" -label:"Persona: Downloader" -label:"Persona: API User" -label:"Persona: Hackathon Participant"

BOOM! There are the open issues I haven't added a persona label to. The goal isn't to obsessively add labels to every issue. The goal is to understand what our community is telling us by opening GitHub issues. I hope you find this interesting. Feedback is very welcome!

Happy hacking!

Phil

* Silly Putty Day is not environmentally friendly and I don't think we'd ever celebrate it. It's just a thought experiment.

quantina-glass-wall.jpg

Philip Durbin

unread,
Oct 2, 2022, 4:17:17 PM10/2/22
to datave...@googlegroups.com
Wow, it's been a while since I've posted to this "issue backlog" thread. I hope everyone is having a good weekend. I spent a fair amount of mine looking at open issues. :)

The last time I wrote there were 681 open issues. Now there are 1253 open issues. (I did close a few.)

One of my main interests is making sure people don't fall through the cracks, especially new people who are opening an issue for the first time. Certainly there were open issues that no one had responded to in over a year and I replied with a "sorry" in these cases.

I spent way less time looking at issues that we opened ourselves. By "we" I mean the team at Harvard and close collaborators. I got curious about the numbers and wrote a little script, byauthor.sh, attached, which downloads issue and pull request data from GitHub and creates byauthor.csv, also attached, to show the number of issues and PRs opened by person. I also made a quick graph of the top authors, also attached. I hope you find it interesting. I'd be happy to talk about it.

Enjoy the rest of your weekend!

Phil

 
byauthor.png
byauthor.csv
byauthor.sh
Reply all
Reply to author
Forward
0 new messages