Any comment about upcoming Git Rev News edition 121 #753
Comments
|
As suggested by @sivaraam in the previous edition, lets gather some statistics to write an analysis on the editions so far and hopefully gather some interesting or fun insights and publish them in an article in this edition. We could gather the following:
Other ideas? |
|
It seems it has been a while there was a video in the "Easy Watching" category (edition 111 ?). So I'd like to plug https://www.youtube.com/watch?v=zigbUJHBsL4. It's made by @JackLot who created the learngit.io course, which was also mentioned in edition 108. GitButler has a steady stream of Git related video's as well. Personally I didn't really enjoy the last two videos (but here are the links anyway: https://www.youtube.com/watch?v=FbW9wlve8sI (released last week) and https://www.youtube.com/watch?v=SdSllNeQuVc (released two weeks ago)). These are fictional Git interview questions, I'm not sure how well that topic fits in Git Rev News? On the other hand, their video about Git Notes was really good. But that video was released in January and is Christmas themed, so I'd feel a bit odd adding it to March's Rev News. I'll keep an eye out for future videos and post back here. |
|
Thanks @To1ne for the suggestions! |
|
Not sure if this is appropriate content for this mailing list, but I've recently wrote a how to guide for new Gerrit Code Review users, could fit in the "Light reading" section? You can find it at [1]. I also don't know if this mailing list is interested in JGit development, but if so I can probably link a few youtube videos on some interesting recent performance optimizations in that space. [1] https://gitenterprise.me/2025/03/10/gerrit-code-review-a-how-to-guide-for-new-users/ |
|
@DanieleSassoli thanks for the suggestions§ @jnareb takes care of the links section, so he will decide on them. Also please note that Git Rev News is a news letter, not really a mailing list. |
|
We might want to write a short article about Usman and Seyi successfully finishing their Outreachy internship. |
Maybe also find out most common sources of articles - looking at the hostname of the site that hosts linked article. Or most prolific cited authors.
Maybe also some statistics, like average and median number of links per edition, splitting into articles (and the like) and tools (and sites). It might be interesting to compute what percentage of links are accompanied with at least one link to previous edition, either as a summary, or even how this number changed with time.
|
|
@sivaraam @jnareb and all, I have started working on such a project: https://gitlab.com/chriscool/git-rev-news-analyzer Let's discuss on that project. Feel free to open issues, MRs, etc. I can give you the right to push to the 'main' branch if you want. (Just give me your GitLab username.) We can also use the TODO.md for small short things we are planning to do. |
|
That's great to hear @chriscool ! I'll try to help with the same when I find some time. My GiLab handle is |
|
@sivaraam thanks, I have invited you as a Maintainer. |
I wonder if mdq: jq for Markdown (stream query language for Markdown) could be used for that purpose. I did not yet look at the tool... |
I'd like to help, too... if I can find some spare time. My GitLab handle is @ |
|
Thanks @jnareb ! I have invited you as a Maintainer too. |
Without slippage, and with single correction 2022-09-31 -> 2022-09-30:
|
|
@jnareb yeah, I think I fixed the 2022-09-31 vs 2022-09-30 issue in the file name in the last few weeks as it failed the stat script, and actually there has been a bit of slippage during the last few years as we often published on the first of second of the next month but the date was still set to the last day of the previous month. If we wanted more accurate stats on this, perhaps we should use the date in the commit that moves the file into the |
|
A bit improved version of the plot:
Here "months" label denote the edition where we switched from "This edition covers what happened during the month of ..." to "... during the months of ...". Found using Edit: is now in the git-rev-news-analyzer repository. |
|
Note that when computing various section-related statistics it would be good to take into account weirdness of the first edition - where some section names and some structure was not yet finalized. |
Done in 4b1f05c. |
|
The article about the 10 years of Git Rev News is not finished. I am still working on the stats. Anyway I will still send the draft very soon as I don't think it's worth delaying this edition further. |
I am slowly working on alternative implementation of the script, using Markdown parser library, in a side branch named |
|
@jnareb, thanks for working on this! And sure, use whatever you want to get the stats you are interested in. As for me, I will take another look soon, but I think I have already covered the following:
I am not sure talking about the "Releases" section is very interesting, but I might take a look at it. Word cloud, or topic detection of articles in "Discussion" section might be interesting, but I am not sure it's possible, without too much work, to do a good job at it. Average number of words, lines, and paragraphs of writeups in "Discussion" section are interesting too, and might not be very difficult, so I might take a look at it too. @jnareb, unless you need help, I will let you work on the links section and related stats, as well as date of publication stats. |
|
Have gone over the draft, interesting edition. Was going to raise PR to add Gerrit 3.12.0-rc0, just wanted to triple check this wasn't automated somehow? |
|
I have added links in acb3547 |
It is indeed automated. @chriscool takes care of this. He might actually run the automation once before publishing the edition which might bring this in. In any case, Christian is a better person to answer this. Speaking about automating, I wonder if it would make sense to have the releases script run at specified intervals of time and have it automatically raise a PR to the repo. How worthy of an improvement would this be ? 🤔 |
|
About statistics: there were 72 interviews / spotlights, @chriscool , not 73 - I think you caught interview from draft, or commented out header in |
I have added some basic stats about the links section in a343b1a. The problem with "date of publication" stats is that for the best results you want figures... and IMHO there is problem with having images in Git Rev News edition - especially because there is version send as an e-mail. Let's drop this, for now. I can maybe create an article about stats from 10 years of Git Rev News, and link to it in the next edition (122). |
|
@DanieleSassoli and @sivaraam yeah, it is mostly automated using get_releases.py. get_gerrit_releases.sh is actually deprecated and I don't use it anymore. It's possible that the script misses releases when things on the release page change or when the site is not available. So thanks for checking. I have run the script again and it found the following releases for Gerrit: 3.10.5, 3.11.2, 3.9.10 So I have updated the Releases section in f6996f5. |
|
@sivaraam having the releases script run at specified intervals of time and having it automatically raise a PR to the repo sounds not worth it to me. There are several issues:
Anyway if we wanted to move towards this we would likely need to improve the script to fully automate the process first. So if someone wants to help with this, please open an issue on the 'getreleases' repo where we can discuss that. |
|
@DanieleSassoli about Gerrit 3.12.0-rc0, the problem with the script and Gerrit "rc" releases is that it would generate |
|
@chriscool don't think it's your problem, 3.12 was supposed to be released yesterday but seems like they only got round to doing the minor versions. Wouldn't worry about it, it'll be included in next month's edition. Sorry for causing confusion. |
@jnareb right, thanks. I did double checked using |
@jnareb yeah, fine. Let me know if there are other things you want to add to this edition though. |
Thanks for the suggestions, @To1ne , and thanks @chriscool for the reminder. Added in 85ec796. |
Thanks @DanieleSassoli for the link. I have added it in 221a761.
Yes, I think those might be interesting (though maybe for the next edition). Thanks in advance. |
|
Edition published and announce email sent: https://lore.kernel.org/git/CAP8UFD0FcsTm70vDrvvOzDuQDEp-N3Sp1PFkP6kLrvxX2XJyXA@mail.gmail.com/ Thanks everyone! |
|
Sorry for chiming in late. It just occurred to me now. I wonder if we should've actually classified Eric. S. Raymond's interview as a community spotlight one too 🤔 Also, I was not able to contribute to the statistics owing to some personal commitments. I'll try to work on the word cloud / topic detection part and hopefully we could use the same in Jakub's article. |
|
@sivaraam no worries for not contributing to the statistics yet. I agree about Eric S. Raymond's interview. It would have been better if we started using "Community Spotlight" earlier for people like him. I don't think it's worth changing old editions or changing the stats we just published in this edition though. |
|
@jnareb I just saw that in the stats you added "Light Reading" seems to be mentioned twice: first: "1090 entries in “Light reading” over 114 editions with 1777 links; around 13.76% of entries mention previous editions." and then: "15 entries in “Light reading” over 12 editions with 31 links; of those, 3 entries mention previous editions." I wonder if there is some typo that we might want to fix? |
The second "Light reading" should be "Easy watching". I have made copy'n'paste error. |
A currently mostly empty draft will be there soon:
https://github.com/git/git.github.io/blob/master/rev_news/drafts/edition-121.md
Feel free to comment in this issue, suggest topics, suggest persons to interview, or use the edit button (that looks like a pen) to edit and create a pull request with the changes you would like.
Let's try to publish this edition around the end of March 2025!
This will be a special edition as we will celebrate !!!10 years of Git Rev News!!! as the first edition was published on March 25, 2015. That's why this issue is created a bit earlier than usual, so we can start preparing it a bit in advance.
Thanks!
cc @jnareb @mjaix @sivaraam @gitster @stepnem
The text was updated successfully, but these errors were encountered: