Whew…Another Summer of Code done with. New experiences, Lessons learnt, a lot of code became mainstream. Overall, an excellent program. So, here are my experiences for this years program:

1.) We participated with ownCloud, Zorp and MATE desktop. I personally think participating with other organizations is a good think and fosters collaborations between the open source communities. It is something we expect to continue in future versions of the program.

2.) As opposed to previous years, all our students managed to complete the program successfully. Their work was praised by the community as a whole. A big Thank You to all the students for their awesome work, and specially to the mentors who keep the wheels rolling.

3.) Me and Manu tried to address the shortcomings of the previous year. We decided on two ‘rules’ for selecting students

– All students must be vouched by a professor (to try and solve the problem of disappearing students).
– All students must send weekly update mails to the opensuse-project mailing list.

The first proposal was met by stiff opposition by some mentors. We even discussed this on the gsoc-mentors mailing list. While well intentioned, I admit that we did not expect many things. We assumed it would be as easy to get a vouch from a professor in Europe and the US as it is to get in India We realised much later that professors don’t have the level of interaction with the students as they do here in India. We scrapped the rule. The original problem is still not solved, and though we have not been hit by it this year, we are open to suggestions on how we can address this problem.

About the weekly mails on the mailing list, I feel it was a good decision. While students may feel that it is excessive ‘red tape’ and totally unnecessary (blogs and project specific ML’s serve well enough), it does help us in the off chance that something goes wrong. We had some experiences in the last edition and felt that it was better to enforce it as a rule. Additionally, it gives community members to provide feedback and improves the visibility of the student’s work.

4.) A lot of work done for openSUSE. Projects such as TSP (Travel Support Program), OSEM (Open Source Event Manager) benefited a lot from the program.

5.) This year, Google is celebrating its 10th anniversary of the program, and having a reunion meet for the same next month. Dominik Bamberger and Artem Chernikov are going to represent openSUSE at the summit.

6.) Mentoring is a tough job and it takes a lot of the mentors time. Henne pointed out that the program is not so rewarding to a mentor, and that we could incentivize mentoring. A lot of ideas were bounced around and we hope to work on them as well.

Lastly, I will be speaking about the Google Summer of Code, and our own experiences at next month’s openSUSE Asia Summit (Yay!). So for this year, over and out

Like every other conference in YAPC too day 3 was far less populated and way quieter.

R Geoffrey Avery (rGeoffrey) - Dancer: Getting to Hello World

A very nice and from the ground up talk from rGeoffrey, what i liked here is him going from the ground up showing the whole stack instead of bluntly talking about how to install run and end a ‘Hello world’ app.pl with dancer. As a small bonus he also showed the differences if you wanted to use mojolicious instead.

Carl Mäsak (masak) - GOTO statement considered awesome

An interesting talk from masak about state, flows and exotic language features.

Susanne Schmidt (Su-Shee) - MVC-templates-routes web frameworks - is that all there is?

Susanne talked about the different approaches frameworks and architectures that are less heard today after being drowned in the MVC fever. Seeing examples from smalltalk,erlang, sinatra, mojo, R (shiny) et al was highly interesting and educating.

Mark Overmeer (markov) - Digest:SHA is broken

Mark showed us how things can break in a module that is supposed to operate on bytes instead of strings (after changing to support strings) even though Perl itself has no way to differentiate between strings and bytes anyway.

Sawyer X - Plack for Fun and Profit (But Mostly Profit)

Seeing how booking.com uses plack in both development and production (with sample code!) was rather interesting, and plack seems to have become a rather nifty tool deserving a good seat in the perl ecosystem.

Upasana Shukla (upasana) - Structured exceptions in Moose

Upasana talked about structured exceptions and the challenges she faced making the transition from string exceptions to structured. She didn’t manage to win me over though with their pros, i still hate (most types of) exceptions and the Object Oriented Everything approach but it was an enjoyable and interesting talk.

Sawyer X - Day 3 Keynote - The Joy In What We Do

For this i will post he’s abstract verbatim: >“Small memories flooding our consciousness like a river of delightful experiences, carrying us through our lives, tugging at our heart, pointing us towards endless appreciation of what many fall prey to forgetting: our love of solving puzzles and our joy of transforming reality and life itself. > >This is the joy in what we do. “

This one was a keynote that deeply touched me and one i would recommend you all to watch and be reminded of the joy in what we do, and the reasons we all contribute to Free software.

Lightning talks Day 2

Being rather tired at this point i only watched one Lightning talk (and i immensely enjoyed it)

Thomas Klausner (domm) - The OAuth2 Dance

In this one OAuth2 was explained through a modern dance choreography that probably confused more than it helped (still was enjoyable and probably tons of fun for everyone :P)

The rest

The rest of the day went buy pretty quiet, the adrenaline levels having dropped with the conference closure, i passed the rest of the day relaxing and getting readyu for the returning trip home to Greece. That concludes the third and final day of the conference and here i would like to say a big thanks to all attendees and especially the organizers for making YAPC::EU 2014 such an awesome conference and a very worthwhile memory.

Today was pretty much a “Devops” day for me, both by watching most of the devops track and holding conversations mostly about/with devops people.

Bradley Andersen (elohmrow) - Regex 101

This was a rather interesting talk in which Bradley went over the bare basics of regex, making an analysis and effective crash course about understanding them and starting to write sane regex. Unfortunately it was too beginner oriented for me but found it quite brilliant for someone who never saw regex before (or as some people like to say “ascii puke”).

Marian Marinov (HackMan) - Create your containers with perl

In his talk Marian, went over the tools he uses to work with containers and perl. In 16 slides, he presented the current state of containers in linux and showed what is covered/supported by perl, he then went on presenting his proposed toolkit full of utilities that combined give you a full environment to work with (in perl).

Naim Shafiyev (shafiev) - Automatization Network infrastructure

Naim, presented the issues an ISP has managing thousands of users with as many different needs and scenarios of use. The current protocols, practices, common pitfalls and especially how to avoid those and learn from previous situations. He closed the talk with some interesting future ideas.

Matt S Trout (mst) - Devops Logique

Matt presented the merits of Logic Programming and taking declarative system descriptions to a new level of abstraction, and finding ways to integrate these ideas back into more common workflows.

Denis Banovic (@bano99) - Perl in the Cloud

This was an interesting talk about running and scaling your own perl application paas on private open source cloud, albeit more dev than ops.

Lightning talks Day 2

Last but not least was the Lighting talks session. A full hour of 10 lighting talks and 30 second breaks between each talk for announcements (or ads).

Laurent Dami (dami) - Virtual tables in Perl for SQLite

Laurent talked about the upcoming Virtual table support in DBD::SQLite.

Kang-min Liu (gugod) - Introduction to Hijk

Hijk, is a very simple very small and very specialized web server (specialized in the sense that it does only one job), also has a hardly pronouncable name).

Simun Kodzoman (sime) - TableEditor

TableEditor is exactly what the name suggests, created on top of dancer and angularJS.

Mihai Pop - Trubzi.com

Brief presentation of Trubzi.com and its features (which of course is built in perl ;)

Dan Muey - I � Unicode

A basic intro to unicode and garbled text issues, Dan also gave a full version of this lighting talk in the past, watching that one is also suggested.

Upasana Shukla (upasana) - GNOME’s Outreach Program for Women

Upasana talked about her experience with GNOME’s Outreach Program for Women & how one can become a part of it.

Matt S Trout (mst) - Day 2 Keynote - State of the Velociraptor

As usual, Matt presented a recap of the previous year in the perl5 community as well how what we saw and learned can help shape both our language and community in the coming year.

The rest

I spent the rest of the conference day talking with and meeting new people. Then went off with the romanian crowd (a bunch of very cheerful,interesting and friendly people) to tour the city of sofia and later go for drinks. That very enjoyable and highly informative sofia city center tour was done by a (native) sofia citizen and also speaker in the conference, Dimitar Mitov (you can find details about his presentation in the previous post.)

And this concludes the second day of the conference.

Two weeks ago, i went to Sofia, Bulgaria but due to being extremely busy with other work these posts got a little delayed. In this series of posts i will describe the 3 days of the conference.

YAPC::EU (short for “Yet Another Perl Conference::Europe”) is the annual conference of the European Perl community. This year the conference took place from August 22nd through the 24th at the Inter Expo Center (IEC). I arrived the day before the conference, just in time to have lunch and head to the pre-conference Welcoming party. The meeting took place to the “Krivoto” Beer House. in the center of sofia.

Next morning I headed with a friend to the vanue, we arrived around 9:30 where they were (still) serving breakfast, there were a few booths. There was a booth with free Coffee, snacks and a couple flavors of Tea. Another booth had traditional bulgarian breakfast, “banitsa and boza”. Banitsa is a filo pastry with cheese and eggs while boza is a fermented wheat kind of drink that is traditionally served with banitsa. During the breakfast time there was a meet and greet session for the conference attendees, a fantastic idea for newcomers both to the community and the conference itself. I didn’t have a camera with me so unless i get my hands on some photos the next sections will probably look a bit bare.

Mihai Pop - Perl Secret

Perl Secret was a funny and really interesting talk from Mihai Pop about perl secret operators and their representation as minions. I like minions nearly as much as i like the secret operators so i attended the talk which paid in full. I had a really enjoyable time hearing about operators i used, remembered or never even heard of, all together with adorable minions!

Mark Overmeer (Markov) - Translation with context

This was a very interesting talk about how to use gettext and perl to have context enabled translations. First he described us how the gettext framework works and how the PO and MO tables are being created. He then went on talking about “Log::Report::Translate::Context” which is a module that using the “msgctxt” field of standard PO-files it adds information to msgids in order to make context sensitive translations possible.

Thomas Klausner (domm) - Code i am proud of

Next came Thomas Klausner who presented as with a very interesting conglomerate of hacks he did last year. The talk included (but was not limited to) Consistent Random Passwords, MOPfuscation, ZeroBlog (which is his current blog’s backend) and a Short and magic $elf (loved that one, was clever :p).

Curtis Poe (Ovid) - How to fake a database design

Ovid’s talk was very good and with lots of tips for you if you don’t want to enrage your DBA. He opened the talk with a series of explanations and definitions of simple (or not so simple) Database design concepts. Then in a small Q & A like session he went on telling us about DO’s and DON’Ts of Database design.

Ervin Ruci (eruci) - A geo search engine in a few lines of perl

Ervin told us about the mod_perl based codebase of a perl module he wrote a few years back. It is the backend powering a Forward and Reverse Geocoding website covering the U.S.A and Canada. He then went on describing how it works and usage examples.

Sue Spence (virtualsue) - Get Me to the Pub Please

In this talk, Sue told us about the local runners group she likes to participate in and the trouble she had co-ordinating with them about when the meetups are, the meeting point, finishing point, the trace of the running etc. She then went on describing what she set up (calendaring, notifications, etc.) and how perl helped her in that (plus the modules she used).

Lightning talks Day 1

Last but not least was the Lighting talks session. A full hour of 10 lighting talks and 30 second breaks between each talk for announcements (or ads). At this point i was quite tired so i only write about only 2 of the 10 lighting talks i remember better.

Dimitar Mitov - Perl Executing Browser

I remember this talk better because i discovered the project a little while before attending YAPC and was quite interested in it. Dimitar opened the talk by telling us why he needed such a tool and the reasons that led him to choose QT as the toolkit for it. In short his browser allows you to have a nice local web interface for your CGI scripts without using a full blown web server.

Alex-P. Natsios - OS-autoinst: Testing with Perl and openCV

And my very own lighting talk about the OpenQA project. OpenQA is the framework used at openSUSE to automatically test the distibution and release media. openQA is almost entirely in Perl (tests included). Due to the limited timeframe i had i didn’t really go deep into how it all works. I gave a fast overview of what it is and how we use it in openSUSE then devoted the rest of my time to explaining how testing works. Finally i showed a real world example with the test for xterm (while explaining its code).

The rest

I spent the rest of the day talking with and meeting new people. I also met old friends and people i had only talked to online (via IRC).

And this concludes the first day of the conference.

As it has taken an extra ordinary amount of time for LSB to move forward a predicament has developed for many distributions, including openSUSE. Many of the requirements for LSB 4.1 can no longer be met and thus while the lsb package still builds it is not installable, see . The technically correct solution would be to drop the lsb package from the distribution, Factory now and thus 13.2 when it is released) as we can no longer be LSB compliant. However, the negative side effect is that some applications we do not and cannot package will no longer install, probably most notably google-chrome. Thus, having no lsb package, or no package that provides “lsb”, is a problem and has negative effects on many users. Therefore, the only way forward is to have an “lsb” package which provides LSB on a best effort basis.

Since LSB 4.1 and previous releases is a monolithic requirement, i.e. if you depend on lsb you depend/get everything that is in the standard, whether you want it or not, it is very likely that many applications depend on lsb while needing only a subset of the features. Thus, while we do not know all of those applications and cannot provide a list of “this will work and that will not“, there is a good chance that many external packages depending on lsb will not only install, but the payload they deliver will work with an lsb best effort approach. For those applications that are broken there is pretty much nothing we can do, sorry.

At a meeting last week at LinuxCon NA the goal was set to have LSB 5.0 released by the end of October. LSB 5.0 is incompatible with LSB 4.1 and prior releases. Thus, even if we provide an lsb 5.0 compliant package in short order after LSB 5.0 is released we still have the same risk as we do with the best effort approach. Basically some application packages that depend on “lsb” will deliver a payload that is broken. Exactly the opposite of what the LSB is trying to achieve, but hey one cannot hang on to Qt3 forever. Therefore, our “lsb best effort” approach to cover the gap does not create any additional pain.

Moving forward the LSB working group decided that the current approach, while providing value, has significant drawbacks. Predominantly there are not enough fingers on the keyboard to continue releases of such extensive “accepted standard” documentation. This is what we presently experience with the non installable lsb package. A for the LSB is being worked out. As this next/new LSB develops we will have to see how application providers deal with this. Without a formal LSB specification in the future, this problem will recur in a few years if application packages depend on a package named “lsb” which at some point may need to seize to exist. We will see how this plays out as the world we create keeps evolving.