Over the course of my master’s research – entitled Bringing Public Budgets Closer to Citizens: An Engagement Challenge for Open Data – some people asked me to share the findings later on. So I would like to use this blog post as an opportunity to write more informally about what I have found and concluded. However, before I get to the findings I would like to provide some background information first.
I came from an open source software background and my interest in the impact of technology in society took me to a master’s program in Digital Culture & Society at King’s College London. In addition, I have always been interested in macroeconomics and its effects on our daily lives. When the time to decide my dissertation topic came, I had no doubt about mixing all those different interests. After some preliminary investigation I decided to research about open budget data, a topic that I was already following closely for a few years. I was already acquainted to this particular area of knowledge due to my involvement with open source software and my master’s module choices were naturally pointing to that direction as well. In fact it was the perfect opportunity to dive deeper.
The research idea came from the following question: are open budget initiatives really capable of improving citizen awareness about government expending and trigger greater engagement on public finances? In order to answer that question, three initiatives that make use of open budget data were chosen as case studies: OpenSpending, an international initiative led by the Open Knowledge Foundation (OKFN) which features a wide community that develops and provides a web infrastructure to publish budgets online; BudgIT, a Nigerian project that comprehends a website, a mobile app and a strong social media presence; Aos Fatos, a Brazilian fact-checking initiative which frequently analyses what politicians state about the country’s budget affairs, checking their speeches against data publicly available and giving a final verdict published on their own website.
Alright, now let’s get to the point! Here it is what the research has found:
Projects such as OpenSpending that aim to provide an easy to use framework through which citizens can publish open budget data still need to lower barriers for contributions. OpenSpeding is built by a technical community of developers and open budget enthusiasts that can potentially engage with the average citizen who is interested in knowing more about their government’s budget. Nonetheless, it is an ongoing challenge to engage citizens who are not academically or professionally involved with public finances.
The OpenSpending Next project, currently under development, is a clear step towards the goal of engaging more people. It will certainly contribute to make open budgets more accessible worldwide by facilitating data upload and sharing.
In order to reach a wider audience, community managers should build a stronger presence on social media. Although OpenSpending has a worldwide reach, the project is not very active on Facebook and Twitter, triggering lower engagement if compared to initiatives such as BudgIT. The Nigerian initiative is highly successful in its communication strategy via social media which is characterized by the use of simple info-graphics and timely status updates.
Access to open budget data in developing countries such as Nigeria is constrained to the portion of population that has access to the Internet. This usually represents a low to moderate penetration potential for open budget data initiatives as the Internet penetration itself is frequently less than 50% of the overall population in developing countries.
The audience of open budget initiatives recognizes the valuable contribution of the information provided also expressing that they would like to see the same content shared on mainstream media in order to amplify the reach of the data disclosed.
In addition, the audience suggests that regional characteristics should be taken into account by the initiatives. For instance by providing info-graphics and information translated into major local dialects.
The availability of open budget data has been very important to the rise of fact-checking initiatives such as Aos Fatos. The diversification of open data sources has enabled faster verification and increased reliability.
Fact-checking projects are seen by their audience as an instrument capable of filtering the huge amount of noise that exists on the Internet regarding the political discourse around certain topics.
One particular feature that singles fact-checking out is the use of journalistic stories along with data. For instance, in addition to an info-graphic, fact-checking initiatives also present a comprehensive contextualization of that particular topic thus telling the story behind the charts.
The journalistic approach towards data contributes to the audience understanding and has impacts on their engagement on social media. However, interviewees reflected that fact-checking by itself is not capable of engaging citizens to put pressure on politicians and decision makers. Such actions are beyond the scope of fact checking thus leaving this kind of practice to other initiatives focused on community mobilization.
Moreover, data gathered from Aos Fatos social media accounts indicate that the audience is very interested in sharing their content but only a small portion of them comment on the original posts. Instead of discussing with a larger audience on the social media official account, people seem to be more interested in sharing the content among their own group of friends or followers thus fostering the discussion between closer people.
In conclusion, the research shows that there are a variety of civil society led initiatives using open budget data using different approaches. However, there are three central conclusions that stand out from the overall analysis of them. Firstly, the organization of communities around open data highlights the fundamental importance of tackling the lack of Internet access in many regions of world so that open data can reach a wider audience. Building data and communication infrastructures is crucial to civic participation. Secondly, governments should heavily invest in data literacy through both, traditional education systems and citizen led initiates such as School of Data. Proper access to education is a must since, in addition to have access to data, citizens should be able to make sense of it. Lastly, initiatives that make use of open budget data are on the right path towards citizen engagement. However, they still need to improve their community and communication strategies to be really capable of engaging citizens at large. The connection between open budget data and the wider audience is fairly recent and it is an ongoing work. Projects and citizens will build and improve openness together. That is the way forward.
For months I have been receiving Facebook email notifications that are not related to my personal account. They are bounded to someone else’s Facebook profile, a namesake of myself. This person’s email registered on Facebook is quite similar to mine. The only difference is a dot between the first name and the surname in the username part of our Gmail addresses. Probably he does not know that Gmail treats email@example.com and firstname.lastname@example.org as the same address as I also did not before running into this issue.
But how bad is that? Many times I tried to prevent Facebook from sending me those emails by using the “not my account” link they provide on them. It was not enough and I kept receiving them. So today I decided to investigate how far this issue could go and used the “Forgot my password” link instead, then Facebook sent me an email with a code for setting up a new password. You can see where this is going. The password was reset without any further verification and I was logged in! I was logged into an account that is not remotely related to me due to an email typo! It seems that the account was active but it was not fully enabled. The owner might have not received the confirmation email, since it was being sent to me. Then I decided to deactivate the account anyway in order to avoid further confusions and told Facebook what was happening by leaving them a message on the deactivation form. I thought that the two email addresses were from different Gmail accounts but investing further I discovered that it does not matter if your email has a dot or not. If you send messages to email@example.com, firstname.lastname@example.org and email@example.com all of them will be directed to the same mailbox.
Anyway, this is a big security flaw. It is not ok that someone can access an account only because other person made a mistake while typing the email address. If the email address is not verified right after the registration then the Facebook account should be deactivated or should expire after some time. I have been receiving this guy’s friends notifications, friendship requests and suggestions for a long time. They should not have sent emails to me at all, besides the first one confirming the email address.
If you feel insecure about this, the best thing you can do is enabling the two factor authentication feature. Facebook will send a text message to your cell phone in order to confirm your identity. This at least will prevent unauthorized access if someone resets your password.
The information disclosed by Edward Snowden to the world created a diplomatic incident between Washington and Brasilia. The Brazilian government is fed up with the United States because it seems that our friends from the north are spying on us. We don’t know the real extent of the espionage endeavors, but the press – notably The Guardian’s reporter Glen Greenwald – is saying that Americans are lurking President Dilma Rouseff, her closest interlocutors and Petrobras, the Brazilian state oil company. But why is it so absurd and what can be done to prevent it?
Let me start by explaining what is not absurd: countries spying on each other. Well, that is just how reality is. All countries have intelligence agencies that, among other things, are responsible for gathering information that can be used to take security measures and also to make economic decisions. The diplomats will deny it, specially the economic espionage. The ones who were spied on will pretend to complain. The spies will pretend to change their methods. It is the diplomacy game. It is well known that states naturally tend to expand behavior and control to outside their boundaries. That is basic lateral pressure theory, commonly used on International Relation studies.
The first real absurd about this story is the fact that an employee, in the position of Snowden, could have access to that kind of piece of information. This is just bad information management and security. It is obvious that the United States government is not protecting highly classified information in a proper way. The second point is that they are not using this information effectively. The huge amount of data the NSA collects doesn’t seem enough to prevent attacks like the unfortunate incident in Boston earlier this year. The last absurdity, and more important in my opinion, is what governments didn’t do to prevent espionage as well as information leaks. The open source software community has been ringing this alarm for years. Jon Hall wrote a nice open letter to President Rousseff about this. Open source should be a crucial element for information and technological sovereignty. States as well as its citizens must know how the software used by public administration works exactly. The code must be auditable. Otherwise we are just asking for trouble. Otherwise we are blind.
I have been working with information security for a while now and I am under the impression that people in general don’t care about security until something really bad happens. Don’t do like our governments. Don’t wait until someone steel your data. Of course there isn’t a system or method 100% secure but risks can be minimized in great degree. We live in an information age and we need to take care of our data, and so does the governments.
This is the hardest blog post in years for me. I’ve seen some admirable folks leaving the Fedora Project during all this time I’ve been a contributor and I always imagined how they felt. It’s my turn, and I know how they felt now.
It was more than 6 years as a Fedora contributor and I’m pretty proud of all of those years, from the yearly days of the Brazilian Portuguese translation team to the two FAmSCo terms, and all other projects: the Spin SIG, internationalization and quality assurance. The Fedora Project is something very important in my life. It is the longest thing I ever did. Longer than high school, longer than college, longer than any job so far. I’m glad that I was able to meet and work with so many interesting and intelligent people from different parts of the word. It was a wonderful experience.
Most of my classmates at college made the choice of doing internships and others decided for scientific initiation scholarships. I decided to contribute to Fedora during college. I still remember how many times I though about stopping to contribute because of finals, papers and classes. Things were pretty hard during the first semesters. Right after those second thoughts something interesting to do appeared at the Fedora land: a new software to translate, a test case that could be improved or a talk at an important event. Frequently I was busy with college related stuff and its many deadlines and Rodrigo always came to me saying something like: “I was invited for writing an article about Fedora to an international magazine. Do you want to write it with me? The deadline is next week!”. Those things were hard to accomplish but they were also fun.
I’m leaving because there are many other things to be done in my life right now. I got my bachelor degree last year and next year I want to go back to study, besides I need to get some rest. Now it’s time to look forward and figure out what else is possible. Of course Fedora will keep powering my computers and open source software will continue to be my main choice. I also realize that there another interesting things going on like open government initiatives and the application of open source principles on other areas beyond software development. I’m definitively going to pay more attention to those subjects in the next years.
I want to say a special “Thank you” to the folks who supported my first contributions to the Fedora project: David Barzilay and Rodrigo Padula. If today Fedora has a growing community in Latin America is because of the groundwork of those two folks. They started out our regional community from scratch, so don’t forget about them. I also want to thank all the Fedora Projects Leaders I had the privilege to meet and work with: Greg DeKoenigsberg, Max Spevack, Paul Frields, Jared Smith and Robyn Bergeron. Thanks for all the support during those years. I’m not forgetting all the fedorian friends who joined the project over the years. I learned a lot from each of you. Feel free to keep in touch! Thanks, everyone!
The 13th International Free Software Forum took place in Porto Alegre, Brazil on July 25-28 and the Fedora Project was there once again. In addition to FISL, we also held two Fedora Activity Days to put translations in shape for Fedora 18.
During our two hour long community event we had several short talks about Fedora. I opened the talks session introducing the audience to the Fedora Project, explaining how our community is organized, our core values and how people can get involved. Jorge Lopes, the coordinator of the pt_BR translation team was next and talked specifically about the L10N project. Then Daniel Bruno, one of the FAmSCo members, presented the features of Fedora 17 and the upcoming Fedora 18 release. After that, Itamar Peixoto, who was awarded with the Fedora Scholarship talked about the ARM architecture and how Fedora currently manages it. Next, Wolnei Junior delivered a talk on Fedora Spins and how to build and customize them. To wrap it up, Hugo Lima presented the RHN Satellite and Spacewalk features. We also took some time to give an interview to the Free Software Radio Station. The chat was about what Fedora stands for and how education can benefit from open source.
The FAD took place at the hotel, so we had the time and space necessary to concentrate on the actual work that needed to be done. We updated the translations for the Fedora website, Anaconda, virt-manager, virt-viewer and setroubleshoot. I am glad that we managed to do a FAD the way it should be done along with other bigger event. Although the main event can be distracting, it is also a great opportunity to get people together and do some work.
I want to thank Daniel Bruno for helping me on the organization of our community event, Neville Cross for his help on the budget side and all folks who attended to our activities and made our participation at FISL happen this year.
It was the my first time I attended to an academic conference and was an interesting experience to talk about the open source way in such kind of event. Another interesting aspect of that conference is its interdisciplinary characteristic, less focused on code and technical aspects, and more focused on management, business, education and the impact of technology on different areas. I believe it was a good conference to publish the paper due to those characteristics and more important than that: an excellent place to talk about the open source way.
Open source is well established in many universities in Brazil. Several universities have labs exclusively to deal with open source software. That is great, but when it comes to community management, process transparency and the open source way of making business the academic sector in general is not aware of the benefits of those approaches. Therefore was nice to have the paper published on a conference like that and to go there and talk about all that exiting stuff. Plus: São Paulo is an awesome city!
As you probably have heard, FAmSCo elections are coming earlier this year. 7 seats will be up for next election and 3 up for the December election. The anticipated elections are a side effect of the FAmSCo election reforms made on the current term. Personally I’d prefer this transition to be made on December, when we usually have FAmSCo elections and when the current term was supposed to end initially. IMHO having elections on the middle of the current term means that the rule is being changed during the game. On previous elections the ambassadors voted for a full term, and my position is that – ideally – we should have fulfilled it in respect to the voters. On the other hand the majority of FAmSCo realize that there are more advantages than disadvantages in making the transition now, for instance the possibility of filling vacant seats.
Although I do not agree with the transition timing, the reform in FAmSCo elections is a great improvement. FAmSCo chair, Christoph Wickert, bravely conducted this effort and now more people will be involved in the elections and the next FAmSCo term hopefully will be better than the previous ones. Despite the initial polemic regarding the transition, the new rules in place are solid and are the result of a work made by people committed to future of our community.
I believe that this was a short but intense term for everyone in FAmSCo and for me it ends with the feeling of mission accomplished. The previous term, which I was also a part of, was completed and more devoted to small but numerous changes, in contrast with those few big changes promoted by the current term. After those two terms I feel like my mission as a FAmSCo member is now fulfilled and I’m not running for reelection again. Therefore I would like to nominate Daniel Bruno as a candidate for next FAmSCo elections. Daniel has been on the road with us for a long while now and has been an excellent mentor for LATAM. He also did a good job maintaining our local infrastructure at projetofedora.org as well as on building up the open source community in northern Brazil.
After the elections I’ll concentrate myself on organizing the Fedora participation at FISL and do some groundwork helping the Brazilian Portuguese Translation Team. The team needs to improve it’s documentation and put some translations back on shape. That’s what we are willing to do on our upcoming Translation FAD and since I have some experience on translation processes from my early days on Fedora I’ll be glad to help them.
First of all I’m writing this post because it can be useful for other countries in Latin America and even other regions like APAC. I realize that countries and regions are different from each other but we can share experiences in order to improve things and make them work better for our ambassadors everywhere.
In Brazil and LATAM in general, we always had problems when producing and shipping Fedora media. The first problem was the budget. Money was hard to get and the process used to take at least a couple months. The second problem, a consequence of the first, was that media was produced too late in the release cycle, sometimes not in time for the events or sometimes shipped using a faster shipment – usually more expensive – so that ambassadors could get them in time. The third problem, was receiving the reimbursement for shipping, what usually happened only in the next release cycle.
An attempt to solve this issue was buying a media duplicator. Unfortunately that didn’t work in the long term. The machine was way too low to copy and print the amount of discs needed, the ink could only be bought in the United States and the maintenance costs were not worth it. Even so, that was a good idea by that time and I supported it. Although not ideal, the media duplicator gave us agility to produce media right before a Fedora release and was appropriated for small events.
Then the machine broke, I was elected for FAmSCo for the first time and fortunately the community credit cards were in place some months after. The community credit card was the first step to get this fixed and I came up with the idea of mass production. For Fedora 15 we tried this for the first time. I sent the original media for production in São Paulo and the company gave us the shipment as courtesy due to the quantity we asked them to produce. A couple weeks after that they shipped the produced media back to me and I started to ship to our ambassadors.
I would like to thank Neville for supporting this as the community credit card holder in LATAM. Without his help that would just not work. This model worked pretty well but we ended with some undesired leftovers. In order to avoid this, for Fedora 16 the Brazilian Ambassadors were asked to post by forehand in a Track ticket how many media they would need during the first six months of the life cycle, so I could base the production on those numbers. I considered a production a bit bigger then they initially asked to make sure we would be able to cover future requests as well and I’m really glad that this time we were able to ship everything and on time.
We still need to work to improve the swag production, but since we figured out the media issue this will be easier now. All Fedora 16 media is now shipped to ambassadors around the country and it is time to start working on quotations for Fedora 17 media.
At this event I will present a paper based on my final graduation monograph entitled “The Effect Of Collaboration On Knowledge Creation And Production Of Goods”. You can check the original paper in Portuguese on a previous blog post I wrote last year or, as you may prefer, the sneak peek written in English.
Although this is not an open source event, the paper is heavily related to the open source way of producing knowledge and software. My own experience being part of Fedora community was crucial to understand how collaboration works in a global level and I am glad that I was able to use Fedora as one of the successful projects studied for the paper. I am really thankful for everything this community taught me and I would like thank the former Fedora Project Leader, Jared Smith, for giving me a short interview for the paper. In addition I would like to thank my employers at Strema for enabling me to attend to CONTECSI and the co-writers for supporting the paper concept.
We are just starting our engines. FLISol, FUDCon Margarita, FISL, and Latinoware are on the horizon as well.
The L10N/I18N Test Week is now over and it’s time to look at whole picture.
Unfortunately I was very sick during the first couple days of the Test Week, but got better in time to run my own test and help others.
I’m glad to see that translators did a wonderful job testing the UI and catching a lot of bugs. Fortunately, Anaconda is in pretty good shape regarding i18n support as well as the overall desktop applications. On
the downside, the lack of a language selector in GDM still breaks the LiveCD experience and LibreOffice langpacks are not being pulled in by yum install libreoffice.
I want thank everyone who helped with ideas and contributions, specially our worldwide community for taking some time to run the tests and share the results.
Don’t forget that we will also be running a Input Method specific Test Day on September, 22nd!
Here are the bugs filed during the test week:
718906 NEW – [te_IN] Disk format message is out of screen during installation
734308 NEW – All the interface strings are not showed in localized ones
734301 NEW – zh_TW] zh_TW does not install the fonts listed on fonts list
734654 NEW – ImportError while installing
734700 NEW – Some string are unlocalized for sosreport message
734298 NEW – abrt has many strings not to be able to translate
734735 NEW – [zh_CN] system-config-rootpassword some unlocalized strings and password tips displayed as question mark.
720240 NEW – [as_IN][gtk][Translation] – Translation Error with “????” appearing
734437 NEW – gnome-utils: the desktop menu for screenshot appears in English
734439 NEW – fedora-release-notes: the desktop menu for the relnotes appears in English
701021 NEW – Translations not loaded when Authencation Dialogue prompted
706756 NEW – No translation on Login-Page of the reboot-menu
726878 NEW – Semi-translated messages (translated strings missing)
734635 NEW – [zh_CN] Redundant question mark in the weak passwords dialog
734643 NEW – [zh_CN] Unlocalized string for authconfig messages
734665 NEW – [zh_CN] im-chooser Unlocalized strings and Gtk-Message: Failed to load module “pk-gtk-module”
734683 NEW – [zh_CN] Some unlocalized string and warning message from system-config-selinux/selinux-polgengui
734717 NEW – [zh-CN] system-config-language the language list is not fully localized.
734303 NEW – Some messages showing in English
734355 NEW – [jp_JP] UnicodeDecodeError [ja-JP]
734434 NEW – Messages isn’t translatable
734283 ASSIGNED – Message showing in English
734709 ASSIGNED – An error message pops up for system-config-boot
734628 ASSIGNED – Some cities and countries are missing in PO file
734387 ASSIGNED – “English” and “To English” isn’t translatable
734656 ASSIGNED – Unable to run with error messages
666038 ASSIGNED – [all languages] yum untranslated words
734309 MODIFIED – im-chooser: Messages showing in English
734305 MODIFIED – [zh_TW] firstboot does not show all the translated strings for zh_TW
734668 MODIFIED – [gu_IN] Missing firstboot translations
734631 ON_QA – The string translated in transifex shows up in English
734338 VERIFIED – translation error in system-config-firewall
734278 CLOSED DUPLICATE – String showing in English
734279 CLOSED DUPLICATE – Message showing in English
734280 CLOSED DUPLICATE – Message showing in English
734302 CLOSED NOTABUG – ‘System’ term missing on the toolbar
734658 CLOSED NOTABUG – Unable to run with error message
734294 CLOSED DUPLICATE – Some strings not translated
734299 CLOSED NEXTRELEASE – mixing up both half-width and double-width characters for parenthesis in authconfig
734314 CLOSED DUPLICATE – “Troubleshoot” in Help menu is not translatable
734723 CLOSED UPSTREAM – [zh-CN] system-config-printer unlocalized string “Settings”.
678583 CLOSED ERRATA – When using pt_BR locale, some tags are not translated like @@NOME_DO_PRODUTO@@
734301 NEW – zh_TW] zh_TW does not install the fonts listed on fonts list
681750 NEW – No Language Selection/Language List in GDM
734976 NEW – libreoffice-langpack-*-* not pulled in by yum install libreoffice