Blog Update: Categories Added

I have added Categories to the blog – there is a list of categories in the panel on the right. I have also classified the previous 250+ blog posts into categories. Currently, each post belongs to a single category. This should make it easier to understand my thinking since its possible to get related posts together. At the end of each post, it gives the Category it belongs it.

Among the next few enhancements we are working on:
– making the monthly archives page into an outline (listing of Titles) so that its possible to see all the entries for the month at a glance
– see if it is possible to link the category at the end of the blog post to the Category page
– enhancing the Search so that it takes you directly to the post rather than the date on which the post appears; also make the search entries show up reverse chronologically (newest first)
– adding a Google Box so I can provide additional links to a topic I may have written about
– adding the Tech Talk archives into the Blog, so that I can reference them individually and they become part of the site search engine also

TECH TALK: Server-based Computing: My Linux Thin Client

For the past two weeks, I have been using a Linux-based Thin Client as my desktop computer. All my files and mails are stored on the server. The only software my desktop resides on a floppy which has the necessary instructions to boot the system up. When my desktop boots up, it first fetches the Linux kernel from the server and then the applications. Instead of Outlook, Microsoft Office and Internet Explorer, my world now consists of Evolution, Open Office and Mozilla. The total cost of the applications: zero.

It is not that I have not had to use MS-Windows I have done so once. This was when I had to send a document to a friend. He uses MS-Word, so I had to use MS-Word to check and make sure that the document I had created using Open Office looked okay under MS-Word. (It did.)

I have been a Windows user for many years. There are still times when I will type up keyboard shortcuts I know for Windows applications on the Linux Thin Client and get some unexpected results. But other than that, it has been a smooth transition. I have been able to migrate my mails of the last few years from Outlook at an IMAP account on the Mail Server. Even the cookies and bookmarks from Internet Explorer were moved easily to Mozilla, thus ensuring that I didnt have to hunt around for the passwords to access some sites where I had registered (and had forgotten the login and password!)

I am using a 3-year-old notebook as my desktop. So far, I was using MS-Windows installed on it along with the applications which resided on the local hard disk (which had, hold your breath, all of 4 GB in disk space!) In fact, the computers age and configuration do not matter I could have used an old Pentium (or even a 486) with 16 MB memory and the performance would have been just the same.

In fact, that was one of the attractions that led me to try out the Linux Thin Client option. Not only could I use an old, inexpensive computer as my desktop, but I could also bring down the cost of software. In the event that my desktop stopped working, I could replace it with another one just as quickly, knowing fully well that since all my data and the applications that I needed were on the server, there would be no loss of data and the downtime would be minimal.

Whats interesting here is not just the Linux Thin Client solution, but the fact that this is a return to computings roots: that of host-based (or server-based) computing. In a way, I am going back to how computing was done a few decades ago with dumb terminals connected to powerful mainframes. As I will elaborate in the coming columns, there are many recent developments which will make Server-based computing a powerful alternative to the Windows-based Thick Desktops of today.

My perspective, as usual, will be how it can make a difference to the emerging markets of the world. Server-based computing goes by the name of Grid Computing in the developed markets of the world. The two have the same basic concepts but are still quite different in the solutions that they offer. And yet, the underlying theme driving both is the same: making computing a utility for the mass market.

Tomorrow: A Brief History of Computing

Emergic Quarter Update

The past quarter has seen us begin and make some progress in our Emergic vision. I’ll review key achivements of the past quarter, and then lay out the plan for the next quarter.

First, an insight into the 3 key innovations which are driving Emergic.

1. Server-based Computing: The Thin Client-Thick Server project is the key to reducing the cost of computing for enterprises. Any old PC can serve as a Linux Thin Client. This way, the cost of the desktop can be brought down to USD 100. All applications run off the server. I think the time is right for emerging enterprises to eschew thick desktops and look at moving processing entirely to the server. Linux and Open Source are making this possible. It has to co-exist with the Windows world. This is now possible with OpenOffice being able to read and write the proprietary MS Office file formats. The Server willreside on theLAN. Think of this as an ASP. Instead of the servers being centralised, they are no distributed to enterprise networks. I’ll be writing more on this in my current Tech Talk series.

2. Digital Dashboard: This is going to be the new desktop for near real-time enterprises. It will be built using Blogs, RSS Syndication, Outliners and Directories. It will integrate information (events) from multiple sources – newsfeeds, blogs internally and externally,and enterprise applications. What isimportant here is for individuals to narrate their work and enterprise applications to stream “events” on to the Enterprise Information Bus.

3. Visual Biz-ic: This is the equivalent of an Integrated Development Environment for business processes (like Visual Basic/Studio for software components). Enterprise Software needs to be architected usingWeb Services and Business Rules for SMEs, and assembled together like Lego blocks. Software needs to be built tobusiness process standards like ebXML and RosettaNet.

Our focus with Emergic is to make computing available to the bottom of the enterprise pyramid for no more than USD 20 ppm (per person month), including hardware, software, support and training. The target audiences are the SMEs in Emerging Markets who cannot afford to pay for technology in dollars. This is where the next 500 million IT users are going to come from.

I have put together separate updates on each of the areas we have been working on: BlogStreet, Digital Dashboard, Thin Client-Thick Server and Enterprise Software.

On our core Messaging front, our collections for the quarter were on target (3% higher than what we had targeted). We’ve made some progress with channels, but not as much as we would have liked. We’ve also worked on putting together a plan for outsourcing some aspects of our Support and Marketing activities. We’ve also been working to strengthen our marketing activities in other parts of India. The next quarter will be challenging: our target is to show a 25% increase in collections from this quarter.

Enterprise Software: Quarter Update

This has been the area where we’ve not had a clear idea of how we want to go ahead. This is partly because its an area where I am personally out of depth. This is one thing I am now working to change. But as we’ve explored various paths, a strategy has slowly emerged.

We want to use PostgreSQL and JBoss as the building blocks. The former is a relational database and the latter is a Linux-based open source application server. This means understanding J2EE and EJB, and also examining some of the IDEs like NetBeans and Eclipse. We’ve also put together a sample application plan for internal use to be built using these components in July.

Our objectives for the quarter here are:
– put in place a roadmap for Visual Biz-ic and the Enterprise Software Components
– also study ebXML and RosettaNet to see how we can leverage the business process standards work
– identify some milestones which we can build on
– for me personally to get greater familiarity with this space
– identify potential partners we can work with

By the end of September, I am hoping to have a much clearer picture on how we intend to proceed here.

Thin Client-Thick Server: Quarter Update

This to me has been the biggest surprise of the quarter. I had not expected great performance from the Thin Clients, and had always thought that they would be for use by the first-time PC users. Well, I have been proved wrong. The solution is good enough even for me, a die-hard computer user, to be using it! I have used a Linux Thin Client for over 2 weeks, with only a single access to Windows. All our Engineering team too has been using the Linux Thin Clients. This has radically “upscaled” my opinion of what the TC-TS project can do for enterprises.

We’ve put together web front-ends and scripts on the server to manage the process of adding new TCs and users. There are still some problems to be resolved like fonts, printing should be WYSIWYG, the occasional crash, zombie processes when a user logs out, etc. But these are within the realm of solvability.

So, now we are beginning to think ahead. In the coming quarter, we want to:
– Move the Marketing team to Linux Thin Clients, and at the same time give them access to Windows applications on demand from their desktop
– Perfect the solution internally. Sometime, in July, we will be ready with a “first-cut product CD” and then we will clean out our existing installation of TC-TS, and reinstall. We will do this till we get everything perfect. This is because before we go out for pilots, we need to have a “repeatable” and clean installation.
– Next, we want to install the TC-TS at 8-10 beta customers in August and September. Take a different set of entities: corporates, schools, engineering colleges, NGOs, buildings (to homes), cybercafes and hotel rooms. We don’t know which would be the best markets to initially aim for, so we are thinking of trying it out in different places and seeing their feedback. We will work with PC sellers / Assemblers / System Integrators in this.
– In parallel, think through the Product Architecture. How can we componentise the different blocks that make up the server? How should we integrate with our Messaging Solution? How can we ensure scalability, such that the solution can work with 30 and 300 users easily? Also, how to ensure upgrades — this is what we are going to have to do multiple times a month since the packages we use will keep evolving.
– One thing which we are also doing is to identify all the potential barriers which would prevent people from switching, and seeing how each of the barriers can be overcome. In particular, its the Windows applications and the MS Office file formats.

So, hopefully, by Sep-end, we will have got initial feedback from some beta customers and have an alpha version of our product ready.

Digital Dashboard: Quarter Update

We began the quarter looking at Jabber and Instant Outlines, and evaluating blogging platforms like Radio and studying OPML and the Google API. We also built an Outline web service, which takes as input a text file and generates either an OPML file or a renderable file with nodes which can be expanded or collapsed. The next step was to create a blogging platform for the enterprise. Somewhere down the line, we hit upon the idea of building an RSS Aggregator. This is what has transformed the vision for what we want to do.

The realisation dawned that the real discontinuity in information processing is RSS. It enables the aggregation and scanning of 10x the information that one was previously capable of. Combined with a blogging platform, an RSS Aggregator can thus work as the base to build a knowledge management system. What would be expected of people is that they narrate the work they are doing, and put together content with comments of what interests them. (The BlogStreet analysers would then build up the relationships between the blogs, and thus the people behind them).

So, as a first step, we have constructed an RSS Aggregator which can post to a Movable Type weblog. I have just begun using this with my blog.

The plan for the coming quarter is:
– Build on the RSS Aggregator making it more robust and feature-rich (eg. showing only the new entries, enabling search on the archives)
– Making the RSS Aggregator work on the server within the Enterprise, thus enabling aggregation of feeds from the outside world and from blogs within. This means, that if there are 5 people subscribing to a site, then the RSS Aggregator will ensure that the feed is got only once. This is important for emerging markets where bandwidth is limited.
– Building a super-RSS Aggregator on the lines of NewsIsFree as a Web Service
– Creating a Blogging platform to allow people to post items of their interest to a personal blog (or a group blog) in specific categories
– Extending to include feeds from other sources (eg. Mail, enterprise software, etc.)
– Making sure that everyone in company starts using a Blog. This means being able to make the Digital Dashboard as a hosted service on the LAN.
– Extending this hosted service to a server on the Internet so we can offer it to others also. By integrating it with BlogStreet, this will offer a seamless application from being able to identify blogs of interest, adding their RSS feeds to the RSS Aggregator, and then blogging items of interest to personal blogs.

BlogStreet: Quarter Update

When we began, we wanted to launch a blog directory and search engine in a month (by May). We started classifying blogs by hand and did about 200, before we realised that this was not the right approach. Blogs are not as easy to classify as web pages. Besides the manual approach would not take us too far. So, then we stepped back and thought through what we wanted to do.

We still want to make a Blog Directory, but this time around the focus is on (a) the hubs (b) identifying blog clusters (c) doing the categorisation automatically. The two building blocks for this are the blogroll (with the blog links) and the blog posts. So, we are focusing first on doing the blog neighbourhood analysis using the blogrolls and links from the top page of a blog. We have built a system using proglets which identifies blogrolls reasonably accurately.

In this quarter, the plan is to do the following:
– Given a blog, identify its neighbourhood, using what we have called the “Commoner” method: take the most common blogs from all the friends blogrolls and give out a most common list of blogs, in addition to myblog friends, as related. That is if a blog appears among the highest number of times in all friendblog’s blogrolls then it is treated as related.
– Give a rank to each of the blogs, and thus be able to identify the top 100 blogs
– Provide a keyword-based Search engine on the top pages of all the blogs we bot
– Link to a blog’s RSS feed (so it can be sent as an input to an RSS Aggregator)
– Automatically classify blogs by identifying the “hub blogs”
– Look within blogs to identify a Blog Post
– Bot the archives so we can build a search engine with the granularity of a Blog Post (ie, if one searches for a keyword, rather than returning just the page on which it appears as search engines would currently do, it can identify the actual blog post)
– Later: build a live map of the Blog Network (one the lines of what Barabazi discusses in his book “Linked”)

The “hidden agenda” is to be able to leverage the work being done on the external world of blogs to organisations to map out relationships within enterprises as their employees start blogging. (To get them to blog is the goal of the Digital Dashboard project.)

1 Billion PCs sold

So says Gartner, quoted in a report in the WSJ:

One billion personal computers have been sold since the advent of the industry about 25 years ago, Gartner Inc. says.

The research firm Monday plans to announce that the milestone was passed in April. Gartner is also predicting the industry can cross the two-billion-unit mark by 2007 or 2008. But doing that will require recovery from a slump of historic proportions.

In the first decline since 1985, PC sales in 2001 fell 4% to 124.9 million units, a result of souring economic conditions, saturation in the industrialized countries and other factors. Martin Reynolds, a Gartner analyst, predicts growth will resume in 2003, in part because corporations that put off upgrading will have to begin replacing their PCs.

Additional demand from China, Latin America and other emerging markets will also drive future growth, he predicts. Potential bumps in the road include slow delivery of high-speed broadband communications and the addition of copyright-protection technologies to PCs that users may not want, Mr. Reynold says.

Adds a News.com report:

“This demand exists because of the power of the PC to leverage intellectual capital, unlocking the capabilities of individuals to succeed and companies to profit,” Gartner analyst Martin Reynolds wrote in a report. “However, expanding the market will require that PCs become smaller and even less expensive than they are today, while delivering greater functionality and performance.”

Some more interesting stats: “Most of the PCs shipped have gone into developed nations. The United States has received 38.8 percent, or 394 million, of PCs shipped. Nearly 25 percent have gone to Europe, while only 11.7 percent have gone into the Asia-Pacific region, the fastest-growing market today…In terms of design, the vast majority of PCs shipped have been desktops. Only 16.4 percent of PCs shipped were notebooks, and only 2.1 percent were PC servers, or servers based around the chip designs originally devised by Intel. These two markets, though, are expected to be the profit centers for manufacturers in the future.”

And then comes the ridiculous:

In India, for instance, one of the largest future markets for the machines, PCs do not sit on desks in open view. Consumers there want to keep more strict boundaries between work and home life, according to Christine Riley, who heads up the People and Practices Research Group at Intel, a small organization of social scientists, designers and anthropologists that studies how humans interact with machines,and the PC is considered a piece of office equipment. As a result, owners wheel them around on tea carts so they can be kept out of living areas. Smaller, less intrusive PCs in this market are inevitable.

Wonder who feeds them these kind of stories.