Older blog entries for roozbeh (starting at number 108)

Just finished reading the Iran-related article that Miguel had posted. Recommended. The very sad thing about Ahmadinejad is that his opponents in the 2005 elections predicted many of the things that now the international media (both the networks and the independents) are saying about him. But well, his second election was as clean as Bush’s second. The voters actually selected him over the slightly-to-the-right-of-center Rafsanjani.

17 Apr 2006 (updated 17 Apr 2006 at 12:43 UTC) »
Speculative fiction: I just read Dan Simmons' April 2006 message. As a result, I am feeling very sick.

It’s hard to read a horror futurological story mentioning several locations in the country I live, and the country’s president. It’s even harder to be reminded of the various rulings in the religion I’ve been born into.

A recommended read, although it clearly mistakes certain things like assuming that Sunni and Shia muslims can easily unite, or there is only one Islam.

Hint: there are several varieties and interpretations of Islam. I just found an Islamic encyclopedia of circa 1000 CE and it mentions 48 different branches of Sunni Islam and 27 different branches of Shia Islam that has existed then. But still, both the reformist and the conservative clerics of Iran belong to the same Shia branch and they have such a wide difference in their interpretations of political Islam. In other words, there may be a billion different kinds of Islam, one for each muslim.

10 Apr 2006 (updated 10 Apr 2006 at 13:08 UTC) »
The Persian calendar: Following my previous post about the Persian calendar, I discovered an interesting property of the pre-1925 Persian calendar. Apparently, before the reform in 1304 AP (1925 CE), not only the length of the months were different from the current regular lengths of six 31-day months followed by five 30-day months followed by a 29-day month (which becomes 30 days in leap years), and not only some of the months were 32 days, but also the length of the months were different in different years.

I should still find what rules they used (astronomical observation of sun’s position? But the spring equinox has the sun in Pisces, not in Aries as the old Persian name is...), but there are interesting patterns. For example, only Gemini, Cancer and Leo could be 32 days, and were 31 days in years that they were not 32 days, and at least one of them and at most two of them was/were 32 days each year (in the table I have, there is one single case when Taurus is also 32 days, while in every other year it’s 31 days, but it may be an error). Also, only Scorpio, Sagittarius, Capricorn, and Aquarius could become 29 days, and they were 30 days in other years. Or Pisces is always 30 days, the only month whose length doesn't change in my table of 1230–1303 AP (about 1851–1925 CE).

Interestingly, the 33-year cycle leap year approximation holds for that seventy-five years, so the most common software implementations out there (used in Jalali, .NET, and Mono), compute the first day of the year correctly for the period of 1851–1925. But since they assume the present month lengths instead of the changing ones, they compute the Persian dates correctly only for about five-ninth (55.55%) of the dates in that period.

Movies: In the long Iranian holidays (and the mood that continues), we watched several movies. Crash was specially nice, and depicted Iranians so accurately (my favorite Iranian part was when after the raid by Anti-Arabs of the Iranian store, when the Iranian mother is cleaning a sprayed sentence amid her daughter’s protests “But couldn’t they know we’re Persians, not Arabs?”) that made me think how much better House of Sand and Fog could have been, but the best of all that really astonished us last night was Emir Kusturica’s Underground. While I would loved to watch this much earlier, I’m not sure I would have appreciated the movie if I knew less about the history of the Balkans...
Fonts: Trivia question: “Dejavu fonts is the best free software font out there.” Right? Wrong!

The story is as short as this: They are now putting everything in. They even claim that “The users want that!” They are following the dark path Freefont went.

But to quote Behdad, “No, that is not users want. Users don't know what fontconfig is, users don't know how font selection is done, users don't know all these things. What users want is that their software works. If the simplest way to give them that for you, the font developer, is to stuff every glyphs somebody contributes into a single bin, this bug is the result, let everyone suffer it.”

The story in detail appears on gnome bug 334758 (continue and read all the comments), where Behdad and Owen, the pango maintainers, are trying to talk sense into the matter. Matthias even tries sticks and then spelling out the terms, but to no avail.

I didn’t know about this until it bit me today: I started to see ugly Persian fonts appear on my Persian GNOME desktop. Running $ fc-list :lang=fa was the answer. DejaVu Sans was there, while it shouldn’t have been. (I was even worried at first that some copyright-ignoring Iranian has contributed the glyphs, as they looked somehow Tahoma-like, but well, apparently they were original contributions by Ben Laenen).

It’s so sad. I just ran # yum remove dejavu-fonts.

18 Mar 2006 (updated 18 Mar 2006 at 14:49 UTC) »
Paul, as a non-mainline distributer who needed to chose between Epiphany and Firefox and has chosen Firefox, I feel obliged to answer why we did so.

The only really important reason was marketing. There is already a huge marketing force behind Firefox everywhere in the world (most of it free evangelism by journalists), it being the main rival of the MS/IE monopoly. With shipping Firefox (and mentioning it in our marketing material), we are building on that marketing wave. So many people have heard about Firefox these days that we will alienate the want-to-be-different users if we don’t ship that and ship some GNOME-y thing called Epiphany that they have not heard about. This also reminds me of a dialog with Anna in Stuttgart. She told me that (some of) the test subjects had no problem associating the firefox logo with a web browser: they had already seen it!

The same reasoning may apply to OpenOffice.org vs AbiWord and Gnumeric.

So in short, if you want a distributor to default to some piece of free software instead of another, apart from making sure that your preference is solid and usable, make sure people (not only the distributor) hear about it. Alternatively, go and help improve the most heard-about software in each category, so it will suck less.

PS: I miss the days that I used Epiphany: I am using Firefox now, because that’s what we are shipping and I am supposed to eat our own dog food.

Freedom of speech: Akbar Ganji (disclaimer: don’t trust Wikipedia on Iranian matters, it’s only good for a rough general idea of the subject) has been released from jail today. While I don’t agree with the guy’s strategy by doing a hunger strike instead of “co-operating”, I’m so glad to see his happy face.

I really believe no one should spend time in jail for his writing, specially when there has been no private complainant. I am not saying this as some idealist or anarchist who believes in freedom or whatever, I’m saying this as someone (like many people in the free software community) who has been abused and insulted several times by idiots or the jealous because of his open work on free software or the wikipedia, and has suffered because of the insults and the abuses. So when I say people should be able to criticize, I also mean these abusers (oh, the line that separates these is so thin).

Let everyone who feels offended come to the court and explain why he feels so, and why should the offender be jailed or fined, instead of getting people arrested because of possible “threats to the national security” or “collaborating with the foreign agencies”. If a panel presentation or an article in a newspaper are threats to Iran’s security, the Iranian government should improve on making the country more resistant to such threats (which may come from people that cannot be imprisoned), instead of putting them in jail.

8 Mar 2006 (updated 8 Mar 2006 at 02:23 UTC) »
Sharif Linux: The whole FarsiWeb team is smashing bugs left and right these days, to get us to a beta release of Sharif Linux, version 2, which will have a full Persian UI experience.

The interesting thing is that the general quality of the patches we create is getting higher each day, and the time to create the patch is getting lower with each bug. Not being able to rely on upstream developers (either Fedora or GNOME for either patches or reviews) actually makes us double check, do more peer reviews, and all that. I’m also looking forward to the day we are going to upstream all the bugs we have found.

Also, using the desktop in the RTL mode all the time, as I have been doing for more than a month now, really helps in finding and cleaning up the important bugs (vs the theoretical ones we've been pursuing previously). The bidi desktop is becoming a reality now, compared to a dream we once had.

It’s been more than six years since kind people like Pablo Saratxaga, Robert Brady, Erik van der Poel, Markus Kuhn, and Owen Taylor helped me enter the GNU/Linux internationalization and localization world. I feel this is getting somewhere after all.

OK, I’m getting over-romantic, but I can’t stop thinking of the people who are helping make this happen: Elnaz, my dear friend and wife who is taking care of both me and the company and is supporting me in all the hard steps, Yahya Tabesh who is our chairman now and has been a mentor and the main sponsor since we started the FarsiWeb project in early 1999, Behdad the once-apprentice who passed the master and has become an uber-hacker many envy, and the many special friend-colleagues, Hamed, Meelad, the two Behnams, Farzaneh, Hooman, Farzad, Alireza, and all the others who came and went. Thank you guys for helping me in realizing the dream. You rock.

22 Feb 2006 (updated 22 Feb 2006 at 22:26 UTC) »

OK, from now on, whenever you see some free software project requiring you to be not in Iran, Iraq, Syria, Cuba and other countries where us terrorists live, or require you to not give the software to terrorists such as us, it would be very helpful if you tell me (roozbeh at gmail dot com). I wish to make sure they know their software is not ‘Open Source’.

To copy clause 5 of The Open Source Definition, “[...] Some countries, including the United States, have export restrictions for certain types of software. An OSD-conformant license may warn licensees of applicable restrictions and remind them that they are obliged to obey the law; however, it may not incorporate such restrictions itself.”

A few free software projects do that (incorporate such restrictions), and have been doing it for a long while. I guess it can’t really get more explicit than what OSD says. I’m planning to go inform them one by one.

We may be from a country whose nationals are automatically considered suspicious for even thinking about visiting the United States (for whatever reason the US administration may have, that may think it can make life harder for the Iranian government by making it harder for Iranian nationals), but we can’t stand our “freedom-loving friends” doing the same to us through their lawyers.

BTW, how many of you have been fingerprinted using black ink, weighted, and photographed when you entered a country just because of your nationality? If yes, did that make you miss your connecting flight which had a three-hour safe margin? If yes, did they also photocopy all your airplane tickets and the boarding pass for the next flight and in the process lost the ticket? If yes, did that require you to stay the night in a city as large as Los Angeles with no hotel reservation and nobody to call in a country that you are visiting for the first time? Still yes? Who are you, North Korean or something?!

Wow! Apparently I share the same Chinese given name with Donald Knuth, one of my personal heroes. It is 德纳, or Dénà in Pinyin. I don't know if Hong knew about this.

(Thanks to Ming Hua, a self-described “Chinese GNOME user” for pointing me to the similarity.)

99 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!