Un proiect al Grupului pentru Software Liber

Caroline is one persona created to illustrate the various categories of Fedora users and she stands for the "Casual User", paraphrasing, she tries to spend as little as possible time on front of the computer but may be involved in some other communities.

Frankly, I used to root for Caroline and want to see her targeted more by our distro, but after seeing the changes in our distro made in the name of catering to her needs... I got to care very little (none?) about her and I am not far from hating her. Why? because those changes are making my life harder and my computer use less pleasant (I acknowledge is not always Caroline's fault, she's sometime used as a strawman to push some agendas).

We look desperate trying to "steal" the Ubuntu audience and forget what made Fedora Fedora and doing it poorly. As I see it, the key to the success on the desktop is: features, features, features, polish, polish, polish. Look at the features we advertise for F14 and, please, show me at least one desktop user that get excited by one of them. Of courde we are not perceived as a desktop distro and losing. Badly.

We at the Fedora Design Team are not arrogant enough to think the default wallpaper we provide is the ultimate choice for everyone, we understand the diversity of our users and their tastes and trying to make a larger audience happy we want to provide a set of alternatives. The idea is not new, we started playing with it a couple of years ago, but only now it came to fruition.

Of course, other distros are doing it too and they may arrived faster to results, but we used the "Fedora way", more Free and more Open :D It started with a wiki page collecting submissions from everyone, it was announced on blogs/Planet, Fedora Weekly News, microblogging and so (as a writer for FWN I made sure the happenings are documented as they go).

After a generous submission interval we had the members of the team expressing their option/preferences/votes (you want the selection made by someone with an understanding about design, usability, photography and such) and this ended in a "winners" page containing the most voted images, some of the submission are creation of Fedora contributors and some are Free images gathered from the internet (Flickr)

The current step is to double-check the licenses (we became quite paranoid about this after the InvinXible/Solar incident) and at the first pass we struck down a couple of them due to restrictive license (CC-BY-ND and CC-BY-NC are not Free, so not acceptable for inclusion on Fedora). The second pass is to contact the unknown photographers and learn if they are happy with the attribution we are able to provide, some already replied positively. Expect the package to hit the repos before the Beta release and get a few improvements until final.

One of the last steps we are still trying to figure is providing attribution (author name, license and URL) in the wallpaper chooser application as tooltip, the current idea is to (ab)use ane XML file in /usr/share/gnome-background-properties/) and another is to try to persuade some spin maintainers into liking and using the package (it will probably be at least part of the Design Suite).

Rock it.

Mulțumiri celor care au contribuit la Kiwi în ultima perioadă:

vnStat este o aplicație mică și eficientă care ajută la monitorizarea consumului de bandă, acolo unde ISP-stul nu pune la dispoziție o astfel statistică sau atunci când se cere o dublă monitorizare.
vnStat rulează în terminal, fără a mânca resurse importante, având și o interfață grafică simplistă cu ajutorul căreia se pot vedea informații grafice ale activității pe placa de rețea selectată.
vnStat nu este un sniffer, informația pe care o analizează fiind cea din /proc. Din acest motiv, vnStat poate funcționa și cu drepturi normale de utilizator, nefiind necesare privilegii de root.

Instalarea vnStat.

vnStat este prezent în pachetele de bază în distribuțiile majore și poate fi instalat folosind managerele de pachete specifice fiecărei distribuții în parte.
Dacă totuși vnStat nu este prezent în pachetele de bază ăle distribuției pe care o folosiți, iaca pașii de instalare:
Se decarcă sursele de aici.
Se dezarhivează:
$ tar -zxvf vnstat-1.10.tar.gz
Apoi se compilează și instalează sursele:


$ ./configure
$ make
$ make install


Odată instalat, vnStat se poate lansa cu comanda:

$ vnstat -u -i interfața utilizată (eth0)

La prima lansare, vnStat își va crea în /var/lib/vnstat baza sa de date în format binar. Pentru obținerea informațiilor ulterioare desre activitatea pe placa de rețea este suficient să rulați doar vnstat.

De asemenea se poate seta o activitate programată în cron care să actualizeze bazele vmstat.

vnStat nu este limitat la monitorizarea unei singure plăci de rețea, acesta fiind capabil să înregistreze și activitatea mai multor plăci configurate.

Pagina de manual vnstat poate fi accesată aici.
Pagina proiectului unde pot fi găsite informații despre versiunile noi cât și detalii de utilizare este: humdi.net/vnstat

…era ca Adi. Fără glume, acu seara eram la cineva și am prins un clip pe VH1: U2 – Where The Streets Have No Name.

Asemănarea e izbitoare :)

Finally, after a hiatus of over a year, the Ubuntu derivative tailored for Romanian and Hungarian Linux beginners is having a new release! Of course English is still available on the CD :)

Last week being out for a photography camp (don't ask for the photos, I don't have any ready to show yet and then will have to wait more for signed release forms) I missed a good part of the talks (mega-flames?) about the Fedora vision and mission, happening on the advisory board, respectively development lists. Only now I start having a little time to get a glimpse of it and maybe making my point in the process.

People that know me pare probably aware I like to quote a lot an old but famous piece by jwz about software development and its purpose, here's an excerpt:

"How will this software get my users laid" should be on the minds of anyone writing social software (and these days, almost all software is social software).

"Social software" is about making it easy for people to do other things that make them happy: meeting, communicating, and hooking up.

I agree it is true about any type of software, not only "social" and to be clear, is not literary about helping people having intercourse but in a larger sense making their lives better.

So I should obviously ask myself if and how using Fedora is impacting my life, here are a few examples:

• one piece of software that changed my life in the past years is Inkscape (I practically re-started doing graphics thanks to it) with the new released 0.48 version, for which we have packages in Koji for F14 and up. Even if it is a mostly bug-fix release, I am obviously anxious to run it but the only possibility for doing so it to jump my desktop to a pre-Alpha F14 and endure a world of pain. I miss the older, better times when the maintainer could be persuaded in pushing such things to Stable.
• as any desktop user I use the web browser a lot and my choice is Firefox here. The new and exciting new release is Firefox 4, currently in Beta 4 and targeted to a final release around the same time with Fedora 14. Unfortunately, we won't have it in F14, so I am using builds from an external repo (thanks Remi!) and endure the pain (for example session restore always brings the wrong version). I miss the older, better times when we used to track in Rawhide the development of applications with major user impact.
• the Instant Messenging client is practically broken for months for me, being unable to exchange files over the Y!M network with models and photographers really worsened the quality of my life. Even while typing this very paragraph I had to (awful coincidence!) decline receiving a photo from one of my nude models.
• as a photographer I use GIMP a lot and when I found a package for the development version in Koji (thanks Luya!) I jumped at it. Still, it is an early development version, probably F15 material (in older, better times the maintainer could have been persuaded into releasing it as as update for F14) with many annoyances. I continue to endure it on the desktop but I had to downgrade to the stable 2.6.x on the laptop when I had to give a GIMP workshop. 2.7.1 is early, but a future development release could be better, I am still pondering on this one.

So drawing a line, is my life better? It does not look so... I miss some older, better times.

PS: can you believe being at a nude photography camp I managed to give a couple of my Fedora business card and struck a few conversations about what Fedora is and why someone should/should not use it with a few fellow photographers? Talking about GIMP there was natural and I found other people using it.

(Da, stiu ca n-am mai scris de mult, sorry).

Uite problema pe care am avut-o azi: sa se ia un script existent care sterge dintr-un maildir mailurile mai vechi decat o data specificata in linia de comanda si sa se ajusteze sa ia si parametru de tip "X days ago".

Sounds pretty easy. And it is. Da' mi s-a facut mila cand am vazut ca scriptul (perl) se uita intr-un maildir cu sute de mii de fisiere si itereaza peste `ls -g -G -U --time +%Y%M%d`, asa ca m-am induplecat si l-am rescris (zic eu) mai elegant.

In primul rand mi-am zis ca e pacat sa faci mega-liste cu ls si ca poti gasi ceva metoda decenta de a face asta direct din perl. Cu ce traversezi frumos structuri de fisiere? Cu File::Find, of course! E in core? Este de la 5.0, nu e nevoie de instalare. Un super feature al lui File::Find e ca exista un programel care se cheama find2perl care accepta (aproape orice fel de) argumente pentru GNU find si returneaza un program perl care face aceeasi chestie cu File::Find (ocazie cu care mi-am adus aminte ca in Higher Order Perl era un capitol foarte dragut cu ce poti face cu stilul asta de callbacks, ar trebui sa-l recitesc). Bonus points ca am descoperit ca stat() face cache la filehandle in _ ca sa poti face repede mai multe teste succesive pe acelasi fisier. Useful trick.

Boon, acum aproape ca avem un script decent, ramane sa-i fac si partea de UI, anume stabilita semnificatia argumentelor. Mi-am adus aminte ca Damien Conway avea niste sfaturi utile in Perl Best Practices, asa ca am recitit capitolul cu pricina (bine ca port dupa mine exemplare electronice pirat ale cartilor din biblioteca), ocazie cu care am adaugat si Getopt::Long in lista de dependinte (don't worry, tot core module). Am adaugat ici-colo niste defaults si niste mesaje, si iaca scriptuletul. Nu e cine stie ce, dar mi-a dezmortit putin mintea (sunt gripat zilele astea) si, sper eu, va fi mai usor de citit/debugat de catre urmasi. Maine fac si niste teste de performanta, sunt curios cum se comporta fata de versiunea dinainte.

PS: Am cautat cateva minute si n-am gasit niste template-uri decente de usage(), help() si eventual POD pt. subrutine. "Decent" inseamna "nu trebuie musai sa fie CPAN-friendly, dar cat sa arate ok la un --help sau perldoc". Please share.

I spent this week finishing up the copr client. It now supports all the functionality that the Copr TG API supports. It’s not much, but it’s a starting point.

I spent a lot of time trying to understand the way repoze.who works and the authentication plugins that we’re using for the python-fedora FAS authentication plugin. I finally understood it, I think… The Fedora client library didn’t support basic HTTP Authentication for TG2 apps so I had to figure out how to integrate that into our authentication plugin. It was quite fun all in all, repoze.who has a very interesting way of doing authentication and writing wsgi middleware is always exciting ;). This patch will hopefully go upstream to python-fedora now.

This next week I’ll probably start working on the buildsystem part of Copr. There are a lot of new things to learn there.