G11N FAD report Day3

The last day of FAD has come and we continued hackathon with some new topics. I have created few languages Workstation live spins for quick testing of Fedora 23 release. The reason we are interested in Language spins is that L10N people can quickly test the Fedora releases. Sometimes it happens that people testing releases don’t want to go through installation so Live media is a good option for them to quickly test their language translations. We are thinking some automation of creating quick language spins using langpacks plugin. Hope to see some work done on this soon.

I also participated in the discussion of determining important packages for L10N with Noriko, Ani and Michelle. I worked on re-prioritizing the list of tier1 packages for upcoming Fedora 24 development cycle. We will soon see its updated on wiki pages and in Zanata as well.

The time went fast in discussions and works we done for last 3 days. Though time had come to close the FAD, we still thought if we could have got some more time to touch few other topics.

We then concluded our last day with every group giving a short summary of their work done in this FAD. I feel really good that by coming together we got some issues resolved, some got progress in their discussions. We also prepared a list of tasks to be done in Fedora 24 development cycle. We are now excited to resume our work on those tasks and finish it as soon as possible.

I would like to thank Fedora Council for approving budget for Globalization FAD. Also, I want to thank Red Hat, Tokyo office for providing us space to host our FAD event. Then I want to thank Jens Petersen, Pravin Satpute and Akira Tagoh-san for doing excellent work in organizing this FAD, without you guys it was not possible to have this FAD. Thanks to all other members who attended this FAD and make this a successful event.

I am leaving tomorrow and hope to meet soon again at some event in future.

G11N FAD report Day2

Welcome to Second day!! Rain welcome us all today. It rained heavily since early morning. We met directly in Red Hat office as it was working day. Today we started with a Hangout with Matthew Miller (FPL) and then we continued with hackathon topics since yesterday.


 I worked with Anish on WordWebEdit. Most work on this is done by Anish. Thanks to him for developing such nice application to correct the words spelling. I helped Anish in testing his application.

It is still under development and not completely ready but soon we will see it complete.


I then continued with Fedora language spins topic. I found that spin-kickstarts has got a bug in it which is preventing generating live media images. I temporarily removed affecting line from kickstart file and used this to generate live media images. There were two solutions with me. One to add manually language packages to fedora-live-workstation-xx.ks file where xx is respective language code I used. Another one to trigger langpacks plugin but that will just install langpacks for their base packages selected already for installation.

Another group had a discussion on Fedora G11N Governance which we really needed now after we formed G11N group. There keeps coming some requests regarding choosing particular font or input method as a preference to some language, some translations related issues. Then organizing events related to G11N activities all over the world. For this we need some committee to track these issues.

More details will be coming soon.

We had Japanese lunch boxes today in our afternoon meal.


Then I had discussion with some other group on improving translations quality. This involved changing the Fedora development cycle milestone of software translation deadline to somewhere post Beta freeze actually. This may help in having most of translations available in Final Freeze releases only. More input/feedback is welcome on its ticket.

I had another discussion on topic language support matrix with Pravin. We thought it will be good to use existing pages and re-align them when we add contents to G11N wiki pages. We will be adding remaining language information on new Fedora G11N wiki soon.

It was a long day for everyone us. We concluded our day with the dinner together.

G11N FAD report Day1

We had our first ever Globalization FAD at the start of this month. We all were excited for this FAD as most of the long time contributors from G11N were attending this FAD. We had a 3 days FAD at Red Hat Tokyo, Japan office.

Because of some meeting room issues we started our FAD on Sunday as first day. As decided we met at Ebisu Station and then walk towards Red Hat office which is close to Ebisu Station. The first day started by Jens Petersen by welcoming all of us. Let me also write here who all we attended this FAD.  We got Noriko Mizumoto and Tomoyuki Kato members from localization (l10n) group. Then we got few members from Internationalization (i18n) group. Then remaining few members from Zanata group. Thus we got around 15 attendees for this FAD on first day.


The first hour we spent on introducing each others and how we will be having FAD for next 3 days.

The first talk was by Pravin Satpute on Fedora Globalization status where he discussed what G11N is, what are the benefits of forming G11N group, activities done by G11N group in last six months(f23 development cycle). Then he discussed the need of having few language experts for every language we want to support in Fedora.


The next talk was by Noriko and Ani Peters on L10N status. They mostly talked about how can we help new contributors by improving wiki pages, a single guide to help them start with L10N, Zanata migration that got completed in Fedora 23 development cycle, then a need to improve string freeze breaking issues.

The next talk was by Akira Tagoh on I18N status where Tagoh-san discussed about what I18N mean, what I18N developers do work (I18N projects) and past activities (F22 and F23 Changes) and bugzilla statistics.


We had a lunch where we got to eat Japanese Pizza. It was a nice lunch. Thanks to Tagoh-san for doing all the lunch arrangements for us in cafeteria.


The last talk was given by Alex on Zanata, a translation platform. He discussed about what Zanata platform is, how people are using it, some statistics on its usage. Then we all had some discussions on Zanata’s contributor license, statistics should be available for every Fedora release per language.

After having status from all the G11N subgroups (l10n, i18n, Zanata) we moved on to Hackathon. We got few groups to work on various topics.

One of the group discussed on Next generation G11n architecture design, another on Fedora Language Spins, another on Fedora G11N Wiki creation/cleanup and last one discussed on Automated Testing for L10N and I18N. I started working on finding a way to have langpacks installed on Workstation Live media. We also had discussions on glibc locales subpackaging.

We then had a hangout with Sirko Kempter at Fedora Cambodian event (Khmer language sprint).

We concluded day with what progress different groups made in their work. It was really a nice, informative, more communicative day for me as I got to meet other members personally, know them well. This will really help me to discuss future things with them easily.

FESCo Elections

What is your background in Fedora? What have you worked on and what are you doing now?
=> I am using Fedora since its first release Fedora Core 1. I joined Fedora as a contributor in Fedora Core 6 development cycle. I have contributed many Internationalization(i18n) packages and sponsored some people in Fedora. 
I have done more than 1600 package reviews in Fedora. I am also a 
provenpackager and helped in fixing packages in mass-rebuilds. I also contributed to few packaging guidelines draft.
    Currently I do package reviews, sponsor people in packager group, 
add new style, new language script fonts in Fedora, maintain some i18n packages. Other than that I like Fedora Applications. Whenever there is any new Fedora Application(like tagger, pkgdb2, fmn) is available or any of its new release, I used to test it and if found any issues, report it upstream. I also test packages in updates-testing and provide feedback in bodhi.

Do you think Fedora should be time based or more feature driven distribution? Or compromise?
=> It should be compromise. Accept Changes that will be ready to be tested by Alpha release and follow the release schedule deadlines.

What are the most pressing issues facing Fedora today (from engineering POV)? What should we do about them?
=> We need to have more testing for Fedora Products and resolve any issues in them. I see we still have some installer, package selection, using dnf instead yum, migration to python3, installing non-default groups in any product issues going on. For some of these issues users need to be aware of these changes in advance by providing them some examples on how the changes will affect 
them and how can they will fix them otherwise their packages remain incompatible in the current development cycle. Everytime such big change comes we endup with filing mass bugs, fixing most of the release blockers but not fully resolve all such bugs. Over the last few releases I saw such leftover bugs remained still open. We should make sure to fix them all.

If we look the development happened in last few releases we can see we got many features/changes development happened in Fedora but all this is not getting properly documented on Fedora wiki. We also need more test cases to be submitted with each Change proposal that people can test on test days. Translations is another thing. Every release we see some translations missed by some packages in Fedora. Sometimes anaconda installer too miss to pull translations. We need developers to also make sure that they will check translation coverage to be 100% for the packages getting tagged in final releases. We need more QA, automation to avoid any last minute schedule slip.

We also occasionally find new contributors asking questions about packager sponsorship. We have been regularly amending the sponsorship guidelines but still there are some questions not answered in guidelines and left to individual sponsor to define. Lack of sponsor for new contributors or lack of response from submitter is one problem. The merge-reviews is another problem that could have easily solved by asking that package group/SIG members to finish those reviews in any Fedora release cycle. But no particular decision on this happened yet.

Care to share a screenshot of your Fedora desktop?
=> I use Gnome as my primary desktop environment.

What are your interests and experience outside of Fedora? What of those things will help you in this role?
=> In the free time I read about mobile technology related articles. I do testing of custom Android ROM's for my old mobile and provide feedback to its developers. I don't think this will help me in my FESCo role.

How can FESCo do a better job communicating with the rest of the Fedora community, or do you feel that FESCo is
already doing well here?
=> FESCo is definitely doing good work. Its weekly meeting logs are always posted on devel list so that contributors can know what is happening in FESCo meetings. But the tickets getting reported to FESCo are not getting lower and the queue is always filled with good number of tickets for each meeting and for future meetings. We need more hands to help FESCo in their work. That does not mean more seats to FESCo but more volunteers to either participate in FESCo meetings to share their views on tickets or on mailing list.

What can you accomplish as part of FESCo that you couldn't accomplish as a contributor to Fedora without
sitting on FESCo?
=> As a contributor to Fedora I can always provide my views on topics in FESCo meetings but as part of FESCo I will try to have Fedora development going forward in the required right direction by providing my vote.

What degree of leeway do you feel that the Working Groups should have to diverge from one another in establishing their own identity?
=> The different Working Groups should use the same existing infrastructure, packages in Fedora. However they can diverge by using certain required features that is necessary for establishing their own identity. I think the per-product configuration will be helpful on how this divergence can be implemented.

How would you define the set of criteria for promoting a spin to a product? What about the reverse?
=> I think spins should continue to stay like we have them currently and I don't think we need to increase our products also. If possible we should work on integrating some spins in our products. The current 3 products are good.
The Workstation product uses Gnome desktop environment. The other desktop environment spins can use the similar PRD to promote them as a product. But, we need to find names for those products then. I don't think we need to go reverse now for already defined products.

With the advent of Fedora Council now, what do you see as the significance of FESCO in Fedora project?
=> I think it's significance will remain the same. FESCo has been 
looking into the Working Group's discussions then the issues like Change discussions, some package development problems, non-responsive maintainers and provenpackager requests. The Fedora Council is not supposed to this work and is a high level decision making governance body.

How "closely" do you, as a member of FESCO, follow the devel mailing list before voting on FESCO meetings?
In other words, apart from your own technical qualifications, what is your typical process in arriving at decisions?
=> Sometimes the discussion on some topic receives many replies on the devel list within a day which takes some time to read and understand what users have to say. But, I will make sure I get enough information about the topic on which voting is going to happen. Before FESCo meeting, I will read the tickets given in agenda, try to reproduce the problem and if I can find some information related to that ticket then I will collect it. Based on this information I can decide to vote.

Anything else voters should know?
=> I work for Red Hat Internationalization team. All other information is already covered in other answers.

Editing pdf file using exiftool

I am really missing pdftk package on Fedora 21. Sometimes I need to modify metadata of pdf file and with pdftk it was easy by using its option dump_data option. With this option you will get a output with various metainformation to modify. This output contains information like

$ pdftk Fedora-20-Installation_Guide-en-US.pdf data_dump
InfoKey: CreationDate
InfoValue: D:20140113225921-07’00’
InfoKey: Author
InfoValue: Fedora Documentation Team
InfoKey: Title
InfoValue: Installation Guide – Installing Fedora 20 on 32 and 64-bit Intel-compatible computers
InfoKey: Creator
InfoValue: DocBook XSL Stylesheets with Apache FOP
InfoKey: Producer
InfoValue: Apache FOP Version SVN tags/fop-1_0
PdfID0: 1882d63af416c1657abf38064d7abc1
PdfID1: 1882d63af416c1657abf38064d7abc1
NumberOfPages: 290
BookmarkTitle: Installation Guide
BookmarkLevel: 1
BookmarkPageNumber: 1
BookmarkTitle: Table of Contents
BookmarkLevel: 1
BookmarkPageNumber: 3
BookmarkTitle: Preface


You can also save this information as

$ pdftk Fedora-20-Installation_Guide-en-US.pdf data_dump output metadata.txt

and when you modify this metadata.txt, you can update the pdf metadata using

$ pdftk Fedora-20-Installation_Guide-en-US.pdf update_info metadata.txt output modified-pdf.pdf

But with the retirement of pdftk package, I started to look some simple option to do this metadata updation and found exiftool a perl package. You can install this package as

yum install perl-Image-ExifTool

To view metadata of pdf file you need to use

$ exiftool input.pdf

To modify pdf metadata you need to use

$ exiftool -Title=”Set/Unset title text here” -Author=”Correct Author name here” input.pdf
I find this very easy to use when you just need to correct the Title of pdf file. The exiftool can be used also to modify image file’s metadata information.



kde-l10n-* added again to fedora-kde-packages.ks

Since last few Fedora releases KDE Live iso images don’t install kde-l10n-* packages due to disk space issue. But then many kde users are demanding to have them on iso image. I see a commit which now again started adding those language packages for KDE iso images. This is really nice for kde users. Let’s hope these packages remain on Fedora 22 Final release iso image.