Linux A-Z

A hugely long article on Linux by Reza Pakdel. Writes Reza: “I would like to note that this is not a spoon feeding article. I am not here to tell you how to install Linux or teach you how to use it in a step-by-step manner. This article is the collection of experiences I have gained in my time using Linux. It is also to encourage you to try Linux and use it in your everyday life.”

ChoiceMail Blocks Spam

Walter Mossberg writes in WSJ:

The spam-fighting program, called ChoiceMail, is being released today by DigiPortal Software, of Parsippany, N.J. It is available through the company’s Web site, at In my tests, it cut my spam to zero.

ChoiceMail doesn’t try to identify spam and block it. Instead, it shifts the burden of effort to the spammers, by requiring them to get your permission to deliver the spam. If they don’t ask for permission, or if you refuse to grant it, their spam messages are blocked and never appear in your mailbox, period.

Here’s how the program works. ChoiceMail examines every e-mail that comes in before it shows up in the inbox of your e-mail program. If the sender is on an approved list, easily created when you install the program, the e-mail immediately passes through. If the sender is on a rejected list, the e-mail is blocked and deleted.

If the sender is on neither list, ChoiceMail automatically sends an e-mail explaining that you are using a “permission-based” system. The e-mail asks the sender to go to a Web page and fill out a permission form. The request is then sent to you for approval. If you accept it, the e-mail is delivered to you. If not, the e-mail is killed.

Spam is problem #1 when it comes to email. ChoiceMail has a simple solution, which should work.

Web Services Orchestration

Writes Jon Udell in InfoWorld:

For years the industry has dreamed of modeling business processes in software and combining them like Tinker Toys. Web services orchestration, the new term for that old idea, becomes more interesting as raw services multiply behind firewalls. But as integration vendors point out, the orchestration layers of the Web services stack aren’t yet baked. The standards pioneers — Microsoft, IBM, and now Sun Microsystems and BEA Systems — are busy in the kitchen.

Two proposed XML grammars for describing the orchestration of Web services — Microsoft’s XLANG, used by BizTalk, and IBM’s WSFL (Web Services Flow Language) — were widely expected to have merged by now into a joint World Wide Web Consortium (W3C) submission. That hasn’t happened. Meanwhile, Sun, BEA, SAP, and Intalio have introduced a third candidate: WSCI (Web Service Choreography Interface). The relationships among these three proposals — and others, including Intalio’s BPML (Business Process Markup Language) and ebXML’s BPSS (Business Process Schema Specification) — are murky.

Oracle’s Collabortation Software

Oracle takes on Microsoft’s Exchange with its new Collaboration Suite. Writes

The new “collaboration” software will allow businesses to manage e-mail, voice mail, schedules, as well as hold Web-based meetings, and will allow employees to sync their information to wireless handheld devices. It is a combination of new and previously released Oracle technology and runs on top of Oracle’s flagship 9i database software.

Also part of the collaboration software is Oracle’s Internet File System (IFS), 2-year-old technology built into the database. The company’s goal for IFS was to make the Windows operating system unnecessary, positioning the technology as a replacement to the Windows File System built into Microsoft’s operating system. IFS stores and manages different kinds of content, including audio, video and e-mail and Microsoft Word and Excel documents. It moves data storage from a PC’s hard drive to back-end servers on networks.

TECH TALK: Server-based Computing: How India can Lead

The solution to building a computing mass market does not lie in either Linux-based PDAs (the Simputer is being launched this month, and expects to sell 50,000 units by 2003 at prices ranging from Rs 10,000 to Rs 23,000 USD 200 to 460) or in low-cost new computers (Via is saying it will launch a multimedia PC for Rs 15,000 and expects to sell 0.5-1 million units in the next year; there is no mention of what, if any, OS they expect to bundle with their PC or if the price includes the monitor). According to me, there is only one way for India to sell 5 million PC units a year: adoption of used PCs (all that is needed is the motherboard with a base configuration of anything better than a 486 with 16 MB RAM, a network card, and keyboard-mouse-monitor) which would cost no more than Rs 7,500 (USD 150).

These Linux Thin Clients, integrated with a LAN-based Linux Thick Server (which can be an existing desktop machine with additional memory and would cost no more than Rs 50,000), are whats needed to open up the mass market and thus create a disruptive innovation in computing.

What the Linux computing environment does is make available, for about Rs 150,000 (USD 3,000) a bundle of 10 Thin Clients and a Thick Server. Each additional Thin Client would cost Rs 7,500 (with an additional Rs 500 for server memory). This suddenly makes it affordable for schools and colleges to set up computer labs. It makes it possible for cable providers to offer homes a bundled solution of a computer, cable modem and unlimited Internet access for no more than Rs 1,500 per month on a rental basis in this case, the Thick Server would be either as a shared resource in the building or at the cable providers office. Shopkeepers can use a Thin Client as a cash register, with the database, accounting software and their product list residing on the shared Thick Server. Cybercafes can now bring down the cost of their computing infrastructure by 70-80% by using server-based computing.

The biggest impact would undoubtedly be on enterprises. Even today, in India, more than 70% of computers are bought by businesses, with about half of those by SMEs. This means business adoption of computers is a measly 1.3 million units a year. In a country where the workforce is estimated to be about 100 million, computer penetration can shoot up dramatically with a consequent increase in productivity if businesses were to take in 5-7 times more PCs. The interesting thing is that this would not entail any new investment from their side the cost of one new computer with legal software equates to deploying 5-7 Thin Clients.

The domino effect of an increase in computer penetration across the board in schools, colleges, businesses and government will create a spiraling demand for software tailored to local languages and business applications. It will create the domestic market we so desperately need to build on our strengths in software. Let China and the other developed markets build the cheap hardware (which we will use 3 years later). Our focus should be to play on our strengths in software, services and other knowledge-based industries. To realise this vision means saying No to Microsoft and New Computers, and re-inventing the future of computing by learning from the past. Will we do it?

Tomorrow: Complementors