Older blog entries for roozbeh (starting at number 27)

1 May 2005 (updated 26 May 2006 at 14:10 UTC) »

Someone asked me what does my first name, "Roozbeh", mean. Well, in Persian, it means prosperity.

But last night I came to the idea that it really means "Guevara": It is the family name of a famous Iranian communist gurilla fighter of the 1970s, Khosrow Roozbeh, who was considered the #1 enemy of the Shah and somehow the Iranian version of Che. Roozbeh was finally arrested and executed by the Pahlavi government, after several unsuccessful attempts, one of which involved changing him with someone who looked very similar to him in the high-security prison. Many Iranian men born in late 1970s and early 1980s are named Roozbeh (I once had two friends, both from the same high school and the same department in the same university as me, both named Roozbeh and friends of each other: imagine the hell when we three met and wanted to talk to each other. No numbering scheme ever worked, of course.), but happily not anyone in the free software world that I know.

And no, I don't understand either Khosrow Roozbeh or Che Guevara. I don't believe I like them either.

Apparently This guy is wondering about the cheap Iranian medicine he got in Iraq, and the commentors are worried about the quality.

What I know from living my whole life in Iran, is that the painkillers, the anti-depressants and such are in very good quality, while the supplements (vitamins and mineral tablets, etc.) are not comparable: the previous generation of the family, my mom, my aunt, ... always ask for certain supplements when I travel abroad for meetings and conferences.

As for patents and Iran, we even have a law for software patents! We couldn't subvert the law process (I've personally tried), but I am very happy to say that not a single software patent has been granted yet (our company's chairman is a member that grants them). The committee is very knowledged, and has not bent to the demands yet. The only thing they are worried about: people filing suits against the committee for not granting them the patents they "deserve"! (One example of a recent patent request: Distributing Newspapers on CDs instead of Paper.)

For other kinds of patents, the process is less restrictive, but more interesting. If you "invent" something, you can go and file the patent, putting all the documentation in a pack. It won't even be read, let alone made public. It will be kept in a safe place, and only used when you sue someone of infringing. If that happens, and it is proved that the infringer knowingly did it reverse-engineering your product, not inventing it independently (and of course that yours is original), you will win the case.

28 Apr 2005 (updated 30 Apr 2005 at 10:29 UTC) »
GUADEC: Sara went to the Lufthansa office, and it seems that their 10-20% discount doesn't apply to Iran in a nice way. If we get the special tourist fare from Lufthansa it will be cheaper than the special GUADEC fare, which is based on their "official" rates. The special tourist fare would be about EUR 525, while the GUADEC fare, considering the discount, would be something like EUR 820. I really hope they are not lying to us.
23 Apr 2005 (updated 4 May 2005 at 14:21 UTC) »
Evolution: Working on a automatic iCalendar timezone generator after I found that there are serious flaws in its timezone database (e.g. for Iran, it doesn't even know about the daylight saving time that was introduced in 1991 and has been in tzdata for ages). It seems that Evolution's timezone database has not been updated for about three years, and even then a very limited MS Outlook 2000 compatible scheme was chosen. This has actually made me miss a CLDR phone meeting (which was scheduled on Los Angeles time), that I wrongly blamed on Lotus Notes (the software used by the organizer).

The problem with timezones also has made me arrive one hour early in a meeting in Tehran, but this time it wasn't Evolution's fault, it was MS Windows/Outlook's fault (used by the meeting organizer), which assumed Iran's daylight saving time starts on early March instead of late March. It has never been so. Microsoft was simply misinformed, and never fixed it because of the long release cycle and the lack of customers in Iran. (I have reported this to Microsoft a few times since about 2001 through several of their "globalization" experts, also providing a timezone switch log to them until 2037.)

I seriously believe this is the fault of the iCalendar standard (RFC 2445) in how it treats timezones. In the minds of the software, I and the organizer lived in two different timezones, both happily labeled "Tehran". This may be really due to its non-recognition of the Olson database, calling it "an informal, public-domain collection of time zone information, which is currently being maintained by volunteer Internet participants" in the meanwhile. That's very possibly because of the politics of Microsoft being involved.

GUADEC: It seems that Lufthansa (the supposedly official airline for GUADEC) staff in Tehran are refusing to acknowledge the special GUADEC fare, telling that "they have not been notified"! Fortunately, Sara, one of the FarsiWeb interns who is also planning to come to Stuttgart, is the daughter of the head of our travel agency. She's supposed to go to the Lufhansa office in person tomorrow and talk to the highest authority she could talk to. If we can't make Lufthansa acknowledge the fare, we'll probably travel with Austrian Airlines (which has good fares but only three weekly flights to Tehran) or Iran Air (generally not recommended unless it's the only way, you're out of money, and an Iranian at the same time, recommended only if you wish to avoid alcohol for a few hours before arriving in Tehran and get your stomatch prepared for the all-kebab offerings of Iranian restaurants), both of which make the schedule a little unbending.

The preparation for the visa goes smoothly. The next obstacle is getting the ownership documents for all real estate one or his/her spouse may own. Elnaz, who is now my lovely wife since late August, has an apartment in her name in Kish where her parents live (in her name only for tax purposes, I assume), which is both fortunate in the way it helps getting the visa (providing more reasons for the visa officer that we would return to Iran after the conference), and unfortunate in the way that it is very hard to get the original document, since Kish is basically an island in the Persian Gulf and we don't have a quick and secure postal service in Iran.

Another colleague who's supposed to come with us, Behnam, is working on his exit visa, because he is supposed to do the military service and needs the approval of the university he's studying at (and temporarily depositing a cheque of about EUR 5,000, signed by someone who would remain in Iran, with his university until he returned he returns) to get it. He will probably ask for yet another invitation letter with his university's name on it as the affiliation (instead of Sharif, where he works), to soothe the university officer into approving it.

It would all depend on us getting the money we are due on time, of course.

Cairo (the one in Egypt): The embassy called on Thursday, almost a week after the conference, telling me that the visa is ready. The officer told me I could use the visa "anyway", if I wanted, something I outright refused. Apparently they had never rejected it, they simply refused to accept it on time for the conference.

Misc: Working on Omega and Aleph. Problems with \nextfakemath for fooling TeX into typesetting the next formula in right-to-left mode. Only necessary because the wholly hacky TeX frequently uses the math mode for typesetting text-only tables, and \nextfakemath not working results in Persian text in tables properly shaped but appearing left-to-right.

My head is practically blowing. I hear the sound of electrons zapping there. The weather applet gives 30°C (86°F) for Tehran. God, I need a cold shower...

This was in mail:

The FSF is releasing a new version of the GPL!  Unfortunately, it's
not version 3 yet.  Instead, it's the third revision of version 2.  Like
the previous revision, it will differ only in that it will contain
FSF's new address.  This version will be available and effective on
Friday, April 29, 2005.
When you next get the chance, please go through your source code and replace the old copy of the GPL with the new version. Also, old notices should be replaced with new ones. The following shell command will do this for you in a sloppy way; its results should be carefully checked:
find . -type f -exec sed -ie 's/59 Temple Place, Suite 330/51 Franklin Street, Fifth Floor/;s/02111-1307/02110-1301/;' {} ';'
We will also be releasing new revisions of the LGPL and the FDL.

Time to update all your sources...

18 Apr 2005 (updated 19 Apr 2005 at 08:04 UTC) »

Quoting Mark Davis of the Unicode Consortium fame for anybody interested in helping in i18n standardization:

We can now generate text file that shows the language -> territory / script
mappings in CLDR in a more readable form than the XML; see
For cross-checking, the end of the file contains the languages / scripts that are not represented. This information is currently in draft state; we'd appreciate it if knowledgeable people looked over the missing items to see what information can be added. For example, here is the first part of the 'unrepresented' scripts list and languages list:
======================================= Scripts Not Represented ======================================= [Bali] Balinese [Batk] Batak [Blis] Blissymbols [Brah] Brahmi [Brai] Braille [Bugi] Buginese [Buhd] Buhid [Cham] Cham [Cirt] Cirth [Copt] Coptic [Cprt] Cypriot ...
======================================= Languages Not Represented ======================================= [aa] Afar [ace] Achinese [ach] Acoli [ada] Adangme [ady] Adyghe [ae] Avestan [afa] Afro-Asiatic (Other) [afh] Afrihili [ak] Akan [akk] Akkadian [ale] Aleut [alg] Algonquian Languages ...
If some of these are in modern use for some languages / territories, please file a new reply on http://www.jtcsv.com/cgibin/locale-bugs?findid=471. Please include both the name and the code of the new relation and whether the language should be starred or not (see the text file). Example format:
Please add: [cop*] Coptic; written in [Copt] Coptic, used in [EG] Egypt and Antarctica [AQ] ..

The data looks to be in a very good shape now. If you have a use for it, you should preferably try the XML version.

You can also send the feedback to me (roozbeh at farsiweb dot info) if you wish someone to review it for you before it gets to the Unicode CLDR committee.

17 Apr 2005 (updated 19 Apr 2005 at 15:37 UTC) »

Awake for almost 24 hours. Thoroughly reviewed the list of correspondence between languages and countries where they are official or commercially important in this (incomplete text version) for the third time in the cycle that started more than three months ago. It seems that me and Mark are finally converging to something good.

Had a company board meeting and the trans-company "free software council" meeting, trying to get the business plan for our customized Linux distribution approved and financed by the VC, but I am not very hopeful, really. Seems that they (the shareholders of the VC, that owns 30% of our company) are more interested in talking about and encouraging entrepreneurship than spending the real money they are making in North America here. Free advice, we're getting, but not the money to implement the same advice.

Put lots of information about Stuttgart on the Wiki page for GUADEC. So tired now, should get some sleep to be in shape for tomorrow's translation party (we'll be doing Fedora this time).

18 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!