Although Sharif Linux comes with a Persian user interface, there is also the option of having a fully English user interface. Even if having a Persian interface is of no interest to you, you can still benefit from the high quality and standard support of Persian text processing. This includes a beautiful standard set of Persian fonts, the ability to create, open, edit, and print MS office documents, the ability to view Persian and bilingual web pages, the ability to open and print PDF files, and a lot more. Also, several settings of the operating system are localized for the requirements of Iran.
Contact us at the email address sales@farsiweb.info or call the phone number +98 21 66025705.
We do not provide Sharif Linux as an internet download, because of several reasons:
You are free to copy the Sharif Linux CDs, provided you follow the licenses included in the CDs and you do not break any law. But FarsiWeb takes no reponsibility for any legal action resulting from this. As far as we know, copying Sharif Linux CDs in Iran is legal, provided that you follow the licenses of the software included in the CDs.
You don't need to pay for Sharif Linux. You can get a copy of the CDs from a friend or a collegue, if he will give it to you for no money.
The two meanings of the English word "free" should not be confused. When someone says that GNU/Linux is free, it means thay you have some level of freedom in copying and modifying it. It doesn't mean that FarsiWeb may not sell the operating system, or that FarsiWeb may not charge for the services it provides to its customers. Many free software companies, including Red Hat sell GNU/Linux distributions like the Red Hat Enterprise Linux.
The Sharif Linux 2 Desktop Edition boxes come with technical support and an introduction and installation manual. If you don't want the manual and the technical support, you may get a copy from a friend. It is free software after all! To quote The Free Software Foundation:
FarsiWeb provides the source code to the packages in Sharif Linux 2. If you want a copy of the source code, please email us at sales@farsiweb.info and provide the postal address to which we should send you the source code. Attach the file on the Sharif Linux CDs called GPL-offer.txt to the email, or include it in the body of your message.
Note: You should pay the cost of the postal or the courier service. (According to the GNU General Public License, section 3b, the charge should be no more than the cost of physically performing source distribution, which includes the charge for the postal service.) FarsiWeb will not charge you for the cost of copying the CDs or buying the physicall CDs and is providing them for gratis, although according to the GPL it may also charge for them.
The free software included in Sharif Linux comes under many different free software and open source licenses. The most common of these licenses are the GNU General Public License (GPL) and the GNU Lesser General Public License (LGPL).
But like every other GNU/Linux distribution, like Red Hat Enterprise Linux, Fedora Core, SuSE, Mandriva, Ubuntu, and Debian, Sharif Linux also includes software in several other licenses. This includes the Mozilla Public License, the BSD License, the X11 License, the Academic Free License, and the Python Software Foundation License, among others.
We do not know of any GNU/Linux distribution that is purely GPL.
Developing, internationalizing, localizing, and testing software is expensive. For this reason, when there was a choice in Sharif Linux between different pieces of free software providing the same functionality, we needed to choose only one. For example, between OpenOffice.org Calc and Gnumeric, we chose Calc because of its better support for Microsoft Office documents.
Between GNOME and KDE, we chose GNOME because it had better support for the Persian language, more than 80% of GNOME was already localized for Persian, we had more expertise in GNOME and its underlying tools and C programming language GNOME is written in, several international free software companies (including Red Hat, Novell, and Sun Microsystems) have chosen GNOME as their default Linux desktop, and finally because GNOME is easier than KDE to use for new computer users. We didn't choose GNOME because of personal preference. We chose GNOME because we thought it is ready for the Iranian users, while KDE still needs work.
We know that the correct and standard English word for the name of the language is "Persian". We know that both ISO and the Iranian Academy of Persian Language and literature prefer "Persian". Because of that, we also call the language "Persian" when discussing it in English.
But "FarsiWeb" is just our name, and we don't translate it to English, like every other proper name, like "Sharif".
Most of the outputs of the FarsiLinux project, as supervised by the AICTC of Sharif University of Technology, have not been useful in developing Sharif Linux. There are various different reasons for this. Sometimes the outputs of a FarsiLinux project were totally unusable to anybody, including us. Sometimes the output is confusing, includes false information, and may mislead the readers or users. Sometimes the output may be useful, but only to people newly introduced to programming or internationalization. Sometimes the output of a project is controversial. And sometimes the output of a project is related to a piece of free software that is not included in Sharif Linux.
The only FarsiLinux projects whose outputs we have used were projects done by FarsiWeb or in one single occasion, FarsiWeb's business partner CyberMehr. These were used in Sharif Linux 2. Sharif Linux 1 does not use them.
For more details, see FarsiLinux output, which is a table of completed FarsiLinux projects we are aware of, and how much we have used their outputs.
We have investigated the problems of the Persian calendar (also known as the Iranian calendar and the Shamsi calendar) thoroughly, and we believe that we currently use an optimum algorithm for handling the Persian calendar.
The Iranian law explicitly mentions that the true solar year should be used, which requires astronomical calculations of the March equinox and the solar apparent noon. The problem arises from the fact that the 1304 AP (1925 CE) law that defines the Persian calendar is ambiguous. The most important problem is that the law does not define the locale to be used for the observation of the apparent noon that is used for determining the first day of Persian years.
This has resulted in some astronomers (including Reingold and Dershowitz) using a Tehran meridian as the locale to predict the first day of the year, and others (including Iraj Malekpour) the 52.5 degrees east meridian (the meridian that defines Iran Standard Time). Because of this, nobody really knows if the Persian 1469 AP will be leap or not. Thus, unless the law is amended to include a locale, any algorithm that claims to work for that year or years beyond that, is assuming a locale that can not be assumed.
Also, before 1304 AP, the length of the Persian months were different from their current lengths. For example, all years included at least one 32-day month. For this reason, all algorithms that assume the current lengths of the months produce wrong results for 1303 AP and the years before that.
The above-mentioned problems practically limit any algorithm (that works without a data table for the number of days in pre-1304 months) to the range 1304-1468 AP.
The calendar implemented in Sharif Linux 2 uses the widely used 33-year arithmetic algorithm. This algorithm matches the official (astronomical) calendar for the whole period of 1304 to 1468 AP. This is also the same algorithm used in two pieces of international software, Microsoft's .NET and Novell's Mono.
It should also be noted that implementing the Persian calendar using the 2820-year arithmetic algorithm, as suggested by Ahmad Birashk and others, is less accurate than the 33-year calendar: first, it fails earlier than the 33-year cycle in matching the official astronomical calendar (first failure is in 2025 CE), and second, the 2820-year suggested rule is based on the mean tropical year, not the mean March equinoctial year.
FarsiWeb is currently working on extending its Persian calendar code to the years before 1304 AP, and has gathered information for the actual Persian calendar used in Iran/Persia in the years 1230-1303 AP. We would appreciate any information about the Persian calendar used in years before that.