Like last year, here’s a list of Debian related goals that I’d like to achieve this year. I might not have the time to implement all the projects, but I like to write them down to keep me motivated. And maybe it can inspire other people to implement some of them (or to help me).
- Finish the translation of the Debian Administrator’s Handbook
The target is to have the book available in April. It would be nice to complete the liberation fund until then so that the book is immediately made available under a DFSG-free license.
- Update the Debian Administrator’s Handbook for Wheezy
- Design and implement the Debian Package Maintenance Hub
It’s an ambitious project that aims to merge and replace the PTS, the DDPO and their respective mail variants. It should also standardize the flow of information directed towards package maintainers. I’m going to use the DEP process to drive this project.
This could easily take most of the year, but hopefully I’ll motivate other people to chime in and help.
- Implement
dpkg --check-dbanddpkg --repair-dbWhile dpkg is fairly reliable, it’s not exempt of bugs and more annoyingly, harddrives/filesystems are not 100% reliable either, thus it happens that some internal database files get corrupted. Given that most files are text based, advanced users can manually fix them but many less skilled users are just left with a broken system that they tend to reinstall.
To avoid this, we could provide a command that would try to automatically bring back the internal database to a sane state by looking for a working backup to restore (while at the same time marking some packages as requiring re-installation since we have some indications that they were present).
- Implement storage of dpkg’s internal files in Git
This would be an extension of the former idea. Installing a package dpkg-db-history (any idea for a better name?) would setup dpkg hooks that would record every database change in a git repository. This repository could then be used to restore the last working version of the database.
Translating the book in English is only the start of the journey. The real challenge is to keep the book up-to-date with each subsequent release of Debian. And Wheezy should hopefully be released in 2012 since the freeze is in June.
Besides those concrete projects, I want to do better than last year on the topic of funding my Debian work. I will thus reiterate some objectives:
- Write useful articles for Debian users and Debian contributors.
They should complete the pages Mastering Debian, Contributing to Debian 101, Debian Packaging Tutorials, and help me increase the audience of this blog.
- Write at least one Debian-related ebook (different from the Debian Admin Handbook) and sell it.
It could be an ebook targetting testing users since I believe that many more users could benefit from it if they had some better knowledge of the limitations and of the way to mitigate the problems that arise from time to time.
Or maybe it could be an ebook for people who want to start contributing to Debian, it could even be bundled with a few hours of mentoring.
- By the end of the year, have at least 1/3 of my time funded by donations and/or earnings of my information products.
This means doing 3,5 times better than in 2011. It should be doable given that the sales of the Debian Administrator’s Handbook will contribute to this goal (once the translation is over).
That makes up lots of challenges for this year. Feel free subscribe to my newsletter to stay up-to-date with my progress and to get my monthly summary of the Debian/Ubuntu news. It’s also a good way to help me reach those goals since you will be informed of all my new projects.
Do you plan to implement something like delta rpm’s?
I’m not sure what you’re referring too. Isn’t debdelta what you are looking for already?
I wasn’t aware something like debdelta exists. It seems it is what I was looking for. Thanks. 🙂
Do you think that DBMail could be a useful component in the Debian Package Maintenance Hub? I’ve been very curious about working with that system to manage email based tasks.
It could be. Although I’m not sure that I would decide to store all mails in the database. My initial gut feeling would be to only index the mails (by message-id or some sort of hash) in the database and keep the emails directly on the filesystem.
I wasn’t planning to provide POP/IMAP either so that part seems overkill for my needs.
Mixing plain text search into relational queries is a nice feature as well. You ever looked at Sphinx?
“Or maybe it could be an ebook for people who want to start contributing to Debian, it could even be bundled with a few hours of mentoring. ”
I would definitely buy it !!!
Hi Raphael,
Another bunch of great projects in sight !
As Lucas, count me in for the “Debian Contributor Starter Bundle”.
Also, git storage of dpkg’s internals my be a nice complement to other great tools leveraging the power of VCSs for sysadmin like etckeeper, blueprint, func-inventory…
Keep on the good work
Bests
“could even be bundled with a few hours of mentoring”
Perhaps you can sell the Administrator’s Handbook together with some mentoring, too.
An example offer could be something like:
5 hard copies of the book, together with 5 hours mentoring, e.g. about getting the maximum out of Debian, for a total of 500 €.
Or sell a special edition of the book at a special price, where most/half of the price is targeted to your Debian related work. This would allow especially people in a corporate by buying a normal book to support your work.
Thanks for the suggestions and the nice comments! I’ll try to follow through some of them this year. It’s definitely encouraging to see that some of you of are interested by such products.
Hi,
Very good idea Christian ! Bundling the book with mentoring could be really helpfull !
Mentoring sessions could also be offered based on the topics covered by the book (or not !).
Finally, mentoring could be done on an individual basis, or in small group sessions so that attendees can share and benefit from others case/experience…
Best
Jé
I’m wondering why you abandoned the xz packaging project.
You said the debian-installer side is complete. And the only thing left would be ensure xz is portable.
I know this requires Python 3 so I wouldn’t be possible for wheezy. But about the next Debian?
I don’t know where that Python 3 requirement comes but in any case, I don’t plant to push this further during the wheezy timeframe. And since Wheezy will not be released before late in the year, it would not make sense to keep it as a goal for 2012.