Linux in the enterprise

Open source business prospects

2009-08-20T14:51:00.002+05:30

I am very happy to demonstrate the viability of a pure Free Software (FOSS) based business model. We work only with FOSS technologies and have been doing pretty well.

We focus on RoR, Liferay, Alfresco and FOSS BPM, workflow (openwferu, jBPM). We also maintain a couple of Amazon deployments.

Our most sought after services are in RoR and Liferay. We have benefited quite a bit from these projects and have contributed back in quite a few ways - code, cash donations, bug reports, documentaton...

We have had to reject a couple of projects based on .NET but never regretted it. The only issue we sometimes face is trying to ensure browser compatibility but otherwise we are pretty well placed.

Though we have released a few FOSS plugins/modules and one product, we intend to release a few more going forward.

STPI registration in Hyderabad

2009-08-20T14:44:00.002+05:30

I recently started STPI regitration for R-Knowsys Technologies. The process is reasonably smooth and the STPI website (http://www.hyd.stpi.in/) gives pretty good guidance. The people there are very helpful too - Ms. Kishori and Mr. Suresh.
Start with the '7 easy steps' document and you can get it done yourself without any consultant.

One thing to note is that you need to be very diligent in ensuring that all the projected revenues and investments are consistent across all the documents.

Once the STPI gives approval, custom bonding is mandatory. This is the one which takes more time. Customs department is notorious for bureaucratic procedures. This is the only step pending for STPI registration.

I will update again once custom bonding is completed.

Unethical registrars blocking domains?

2009-06-03T13:28:00.004+05:30

I have been searching for a domain name for a new service we were planning to launch. I shortlisted a few good ones which were available and went back to register them the next week. To my surprise, I found them not available. These are not common names - they are very specific to a niche service and I doubted that somebody else would have thought the of same names at the exact same time.

On googling, I found mentioned on various sites/forums that 'Network Solutions' and some other registrars engage in something called 'domain kiting', 'domain tasting' and 'front-running' where they basically block all domain names that were searched for and force you to pay more.

There is a work around to this -
http://instantdomainsearch.com/
http://www.saferwhois.com/

Hope this helps somebody.

Microsoft threatening to 'send people over' for audit

2008-06-16T16:37:00.004+05:30

I just got a call from Microsoft asking me to purchase licenses for SqlServer and .NET

When I informed them that we were a Free Software development company and did NOT use Microsoft products, the lady - a Ms. Maninder (if I remember right) told me that 'that was not what she had in the database' and that I had to buy licenses. When I refused to do so on account that I did not need them, she threatened to 'send people over'.
Any person who takes the time out to go through our web-site will understand perfectly that we do not do any kind of Microsoft work. I am starting to think that we are being targeted because we are a FOSS company.

Now I am figuring out how to get legal help to fight this. I have no intention to allow any external people onto our premises.

Abolishing Software Patents - Throwing the baby out with the bathwater?

2008-05-27T13:56:00.007+05:30

There is a strong movement in the Free Software Community to abolish patents. Here is a perspective from a real, true blue *small guy*.

When I started my IT career with the then biggest Software firm in India in 1997, my salary was about US$ 1500/annum in India - yes there is no zero missing. It was Fifteen Hundred dollars per annum. It was a meager amount even by Indian (and in general 3rd world) standards. I had to scrape around for money for gasoline for my motorbike.

I worked with other firms, eventually got promoted to being overpaid when I decided to start a firm based purely on Free software (later branched to cover FOSS software). I started with US$ 300, a borrowed laptop and dial-up Internet access. Actually I could not look at starting a firm with proprietary technologies as the license costs were too steep - remember that we are in India.

I really believe in FOSS and think it is great for unlocking the productivity of 3rd world countries. I am convinced enough to bet my life (not to mention my savings, career and marital bliss) on starting a FOSS consultancy. We are doing pretty well and have also contributed back to the community. In spite of that, I am not convinced with the argument of abolishing patents altogether. The purpose of patents (and copyright) was to benefit the society and reward the inventor. The fact that the current patent law is being abused by large firms is a reason for modifying the law - not for abolishing it altogether. We have to ensure that the laws are changed to serve the original intent of fostering innovation to benefit society and rewarding the inventor.
Abolishing patent protection for software altogether will just enable established large organisations to further enhance their monopolies. The small firms where quite a bit of innovation happens will be left with no protection from large organisations which can absorb the innovation (for free) and then use their marketing/marketshare muscle to simply put the small firms out of business. Apart from people who write software for the love of it (and are good at it), small firms and individual inventors will not have much of a incentive to innovate.

Take the case of firms like mine with less than 20 people. If we had hit upon a innovative idea and decided to implement it as FOSS without patenting it, any large corporation can just re-implement it in their products and we will be left with no differentiating factor.
In fact, no matter how hard we try and provide value to our customers, the large corporations can just let us make all the mistakes and simply adopt all our successes. There will be no reason for a customer to choose our product/services over the established players. After all, the bigger players can demonstrate more stability and bigger support infrastructure. To top it all, the large players need not open their source code as they are not using my code - just my idea/concept which they have implemented in their proprietary product.
There is also no reason for the large players to 'play nice' with small firms.

In this scenario, our only hope for survival is if we can get some kind of legal protection. If indeed we are awarded a patent, the larger established players will be comp to license it from us and thus small firms like mine will be compensated.

If the large cos. are abusing this system to create barriers, then we have to look at preventing the abuse rather than abolish the system itself. Just because somebody is into credit card fraud does not mean we abolish credit cards altogether.
In India, there was a 10 year tax holiday for software firms (which ends in about a yr.). There were other firms like realty, manufacturing etc which registered themselves as software firms to benefit from this. The Government should look at tax fraud instead of abolishing the policy altogether - That would be a knee jerk reaction and simplistic.

The government is expected to provide a level playing field for all players. If the small firms do not have patent protection, it will be difficult indeed for small firms to be founded much less succeed unless as a result gross incompetence on behalf of the large firms.

I would recommend a more democratic solution to the software patent issue:
1. Let patents be granted only to individuals and small firms(Criteria - can be discussed).
2. Even in large firms, patents will be granted to individuals/small teams. The large firms should not be allowed to restrict independent licensing of these patents. They can at the most ask that they get exclusive access to these patents for 6 months from date of filing or to recover the cost of the patent.
3. Free patent organisation, Apache foundation, GNU can also patent stuff which can be leveraged to open up more source code.
4. GPLed (GPL v3, Affero GPL) software can be made exempt from patent prosecution if the software for the complete(and not just partial) product/service is GPLv3 or under Affero GPL. This exemption will serve the primary purpose of benefiting society (with patents being a vehicle to promote this).

This way, the society is benefited from innovation without the inventor being shortchanged.

Government liasoning for startups in India

2008-01-19T15:13:00.001+05:30

Do NOT deal with any govt' agency on your own. Always deal through your lawyer or Chartered Accountant or through a competent qualified professional.

In lots of cases, the officials themselves will not know the correct procedure or law. If you go directly, you will be misled and/or misinformed. Always deal through a competent professional.

orgChart plugin for Rails

2007-11-22T11:31:00.000+05:30

Quite a few of our applications require capturing and using the organisational Chart of our clients. We finally decided to make a plugin out of it. We have launched the alpha version of it on rubyforge. It does need improvements and we will add features as and when we require them in our appplications - meanwhile it will be great if people start using it and report issues/feature requests. http://rubyforge.org/projects/orgchartplugin/

Issues with partnerships when starting ventures.

2007-10-25T19:46:00.001+05:30

As an entrepreneur, I have had quite a few offers where people come to me with ideas that they have come up with. Some are quite good actually and some not. People expect me to fund these ideas and take them to market. Rknowsys sometimes does help people with this. I have had quite a few unique problems in this process. Our usual modus operandi is that we first discuss the idea to see whether it does have monetary potential. After that we look at logistics and issues involved in taking it to market. We work with some management consultants. We also work with i-runway.com for Intellectual property issues. After that comes the technology part which is where R-Knowsys is paid either with cash or stake in the new venture i.e. R-Knowsys acts as an incubator. We have relatively easy access to VCs due to our partnerships.

Now, a lot of people underestimate the effort and difficulties involved with entrepreneurship. We need a very well defined product/service offering. We need clarity on what we are NOT doing and at some point people need to be paid for services - including R-Knowsys built prototypes. Recently I have had 3-4 people come to me with ideas. One of the ideas was a pretty good one too. My friend wanted to have a web-site dedicated to educating consumers for a very specialised financial product. He wanted to make money off advertisements. Not a bad idea but I wasn't sure if people will visit the site just for education. I pitched the idea of comparing different products in terms of features and cost and then actually selling the product online. We got pretty excited and R-Knowsys built a prototype and we started going about making the business plan. We made some initial progress but then we started facing issues. Eventually my friend landed a very high paying job and risking that for this venture did not appeal to him. He lost interest and dropped out. There were a few issues which could have been handled better. We had a deal with various vendors. I had paid some of them and it was understood that my friend will pay for one vendor. The liability was split roughly based on stake in the venture. One of the vendor's work was slightly short of our expectations. My friend resisted paying this vendor and I ended up picking up the tab as I have continuing relations with this vendor. I think that this is a very important decision that should be made upfront. What happens to liabilities if the venture is abandoned midway?
The second issue is that R-Knowsys is sort-of expected to put in effort for prototypes/solutions and valuation of this work could be more professional. For Eg. Lets say we put in 3-4 man months of effort for a working prototype which we demonstrate to a VC. What is the valuation of this? Now, we are a service company and charge in dollars/hr for our work. We are based in India and though we pay on par with the market, this is lesser than the actual earnings. Now I believe that the valuation should be what we would charge any client for this. i.e. It should be around USD15k for a 3-4 man month effort. We have had people looking to value our contributions based on our cost rather that opportunity cost. We are working on correcting this, the time/effort we spend on these projs should be considered as lost opportunity with respect to our services business. This has been one of our bigger challenges.

Other issues include - lack of interest/drive from some parties. This is a bigger problem than it appears. There needs to be a clear agreements on responsibilities even among founders. Though painful, formal legal agreements even during prototype stage are a good idea. I have seen parties who are extremely risk/responsibility averse but hang around just in case things work out.

Hope to discuss other issues as we come across them.

Launched RailsSecure on rubyforge

2007-10-25T19:38:00.000+05:30

We had identified quite a few components needed for enterprise applications. Today we launched another Rails plugin useful in collaborative web applications. Normally when we upload files to web applications, they are stored in unsecure format on the server. We launched a new Rails plugin which will encrypt and store uploaded documents - this will keep the data safe even from system administrators and application administrators.

Link to the plugin - http://rubyforge.org/projects/railssecure/

More plugins to come.

Interest from student community on Open Source projects

2007-09-15T17:13:00.000+05:30

In response to the mail asking for volunteers for our GPLed products, I had a couple of students mail me about wanting to contribute to opensource projects. I have an idea of going round undergrad and grad schools in the area and conducting workshops to encourage students to contribute to open-source.

I will float the idea in the local LUGs and see whether there are any takers. We can help solve one of the biggest reasons for non deployment of open-source in India - scarce availability of skilled coders and system-administrators.

Hopefully this can be extended to make a self sustaining project where undergrads from engineering schools start contributing to opensource when they are in still in school and thus build credentials to be showcased when they are searching for employment.

R-Knowsys will start a program to do this. I think we should have it up on our web-site too.

First opensource release from R-Knowsys (and other planned)

2007-09-15T17:03:00.000+05:30

This is the first collaborative product from R-Knowsys technologies - http://rubyforge.org/frs/download.php/25301/ideaxchange-0.1-alpha.tar.gz.
We released the sources a week back, made some minor changes and released the first version for testing. We have tested it throughly internally. There are no known bugs.

We have kicked off a couple of other projects -

VOIP plugin for Firefox - https://sourceforge.net/projects/fvoip.
Distro aiming to be a communication server.

We have some more ideas;

After we have stabilized the ones we have launched now.

Security framework(plugin) for web apps - securely sharing artifacts among community members - the way we share google-docs. A few additional features are planned. This will be a generic architecture and the initial implementation will be for Rails (RoR). We might also implement it in JAVA/J2EE to be usable in other frameworks/apps - Liferay/sakai.
Workflow framework(plugin) - A generic workflow plugin architecture. We will provide tools like creating and editing workflows and ability to plug into new apps with minimal coding. Initial implementation will be for Rails (RoR).
Office applications in the browser - We can build on the Security and Workflow framework above to build a full fledged browser based Enterprise Office Suite. We might need to include version control with subversion/cvs. This will be very useful for SMEs in India.

I have another thought: I have been mulling on this for almost 2-3 yrs now.
We should open up the test cases and show the results of testing for every release. That way people can contribute test cases as well as go through the test results and check for themselves what is broken and how good the release is for production use. Users can also check for themselves the quality of testing the product has undergone. The idea is to have replicate the distributed development model we have right now to a distributed testing model too.
This will go a long way in assuring people that they have an 'Enterprise Quality' product.
We already have excellent bug reporting tools. We may need an open test cases repository. This could be another project for the open source community.
I will try and explore more on this and see if it is feasible to do this for this project.

Please evaluate our first attempt to initiate a free product:
http://rubyforge.org/projects/ideaxchange/

Volunteers welcome.

OpenSolaris vs. Linux

2007-09-11T11:50:00.000+05:30

Have been following this story for quite some time and knew even 3 yrs back that Solaris WILL have to be GPLed for it to compete with Linux. I seriously think this is a bit too late. While Solaris definitely has better features, GPL v2 would have been suicide as Linux is on an upswing and all these features would have been put into Linux faster that you could have blinked (well maybe inside of a year).
Going with v3 is a very smart move considering the Linux may stay with GPL v2 and thus unable to use Solaris code. This strategy is fraught with the risk that other hardware/software vendors may not play ball. It will be easier to put in binary blobs drivers and these companies can then be convinced that freeing up the code may be a good thing. Take ATI for example. GPL v2 was flexible in that respect. What does v3 say about this?
SUN will have to show excellent value to their vendors for the vendors to open up.
While I really welcome this move, I suspect uptake with the community will be slow due to reasons like comfort factor, established user and developer base, packaging ease (livna, rpmfind.net, rpm, apt-get, no of packagers) and given that Linux is still viewed as a community while Solaris is still a 'SUN sponsored project'.
The 2 big challenges for OpenSolaris are:

Vendor buy in.
Building and nurturing an open, motivated and thriving community.

and I suspect thte the second one is going to be tougher. Only if SUN had done this 3 yrs back when it had much clearer advantage over Linux.......I saw the writing on the wall moved from Solaris to Linux fulltime in 2000.
I am starting to like SUN more and more with their strategy on Java and OpenSolaris.
Will either SUN or an independent company work on Solaris on the desktop and handhelds? I do not see any obstacle here. While hardware cos. like DELL, HP, may look at Linux with suspicion/scepticism, they will definitely be more favorable to an OS with a company like SUN behind it providing support and indemnification. Opensource is sure getting interesting as we go along.

Will this turn out to be GPL-v3 vs. GPL-v2? It will be a great benefit to end customers in the long run. And GPLv3 does indeed have some practical drawbacks over v2 and is superior in some respects.

Need to give openSolaris a thorough try out and see if my clients like it better that Linux if they want to migrate their environments. Will have to check out the value proposition in more detail.

One thing is for sure, R-Knowsys WILL launch a product for OpenSolaris under GPLv3.

Enterprise products in Linux to beat MS Office/Sharepoint

2007-08-27T19:51:00.000+05:30

The most common applications in enterprises apart from communications are office suites(word processing, spread sheets) and collaboration/workflow.

These needs are currently being met by MS-Office product range and either MS-Sharepoint/Groove or other proprietary/custom developed solutions.

There are a couple of strategies for Linux to provide value here:

Openoffice + Portal/CMS (Liferay/Alfresco).
Completely browser based solution.

I am not a big fan of solution '1' for the enterprise. The management overhead is too high if you want to manage Openoffice (any desktop application for that matter) on 1000+desktops. I will not discuss too much on this.
We have envisioned a more scalable and easier solution. What we are working on is:
Browser based office solutions(like google docs) with the following features:

Browser based editing (store docs on server as xml)
Workflow engine and workflow features (Modular, Generic, Customisable for each enterprise, multiple workflows as required. User/Admin created workflows 'on the fly', Standard doc templates and workflow templates.)
Security by data encryption - Ability to encrypt any data by user and ability of user to share encrypted data without exchanging keys - transparent sharing of encrypted data. protects data from sysadmins too
Version control (using subversion) - This will be a key feature and easily implementable on a server based system
Additional collaborative features like integrating IM/voice/video with the docs/artifacts.

Example scenario:

Marketing executive (exec) creates a new document based on an enquiry from a client(Enq-doc-1). He does this by filling in an on-line template(Enq-doc-template) standardised within the organisation. He fills this template over 2-3 days saving drafts whenever he puts in more information. Once he thinks he has filled in all information, he saves it and clicks on the 'submit' button.

The document is automatically routed to his boss - the branch manager (br-mgr). The boss reviews the document, thinks that some more information is needed.
The boss asks for the information in either of these ways:

Types requests in relevant section of document itself and 'send back' to executive.
Types in the requests in a separate 'comments' text box provided in the browser.
Puts in an IM message.
'Talks' her message into the browser and 'sends' it to the exec.
This message is by default 'attached' to the context of this document. (As an option, the message may be set in global context - i.e. a general message not related to this Enq-doc-1)

The exec receives the message and fills in the required information. He 'talks' into the browser to say that the client is very keen to sign the document ASAP. This voice recording is again saved with the context of the Enq-doc-1 and can be listened to by the br-mgr.
Normally the br-mgr and provide the required input which is routed to the exec for him to provide the quote to the client.
In this special case, the order is a big one and the br-mgr decides to take the support of the regional HQ mgr (rHQ-mgr) before supporting the quotation. She verifies the document and 'dynamically' changes workflow to forward it to rHQ-mgr sitting in a separate office. They arrange to have a VOIP conference call where the rHQ-Mgr and br-mgr get together with the exec and thrash out the details. The exec updates the quotation, gets approval from his boss and the rHQ-mgr and mails it out to the client.
He visits the client and some more negotiation comes up. He immediately logs into the Intranet portal and arranges an Video conference between the client, br-mgr and rHQ-mgr. The rHQ-mgr offers a slightly changed solution which the client likes and confirms the deal. BTW the video and audio of the conference is automatically saved in the context of Enq-doc-1 (and can be optionally deleted if storage is a problem.)

Key points:

All interaction is browser based.
All context is saved.
The most convenient medium of communication can be used.
No information is lost and mis-understanding can me minimised.
We have every version of every document.
Document templates
Workflow templates
Document to template mappings 'on the fly'.
Multimedia capabilities.
We can explore whether we can give this as an installable at a later stage.

It might be a good idea to have this as an independent distro. This product is aimed at thin client operation.
Sub-projects here will be:

Extensible, modular Workflow engine.
Ruby security/encryption plugin.
Portlet interface to previously discussed communication server.
Browser based Office applications.
Generic export/import for various Office doc formats and PDF.

Proposed Enterprise linux altenatives to MS products - MS-Exchange

2007-08-27T17:13:00.000+05:30

Date: 27-Aug-07

Exchange is easily one of the darlings of enterprise IT Managers. Its main advantages over opensource equivalents (as a solution - not as a single product) are:

Much easier to administer for a large number of users with different requirements in an enterprise.
Mail/Calender client integration.
Delegation of powers of administration.
Delegation of email access (to secretaries of busy managers etc.)
Real good integration with ADS/Directory server and Windows in general.

What features are necessary for an opensource solution to provide excellent value to customers?
Instead of looking at individual mail servers and calender applications, let us try and look at what solution can we provide with opensource.
The way I see it, we can provide a complete communications solution to enterprises. The various components can be:

eMail
Instant Messaging
VOIP
Video conferencing
Collaboration halls (Like Internet chat rooms.)
Maybe calendering
Groupware suites (I have my reservations here. E-groupware looks like a good candidate but it doesn't seem to have that 'polish'.)

From our experiences, we have decided to build this product for our customers. Our proposed solution:

Make it an independent distro - That way we will not have installation dependency issues and this product can be deployed on an independent dedicated/virtualised server. We can control external risks as well as provide a comprehensive admin UI without having to worry about other Linux components.
Configure all the above mentioned servers during setup - i.e. ask all the relevant details like IP addresses, mail domain, IM domain....during setup and configure them so that after setup, we have fully functional communication server without having to configuring individual servers after installation.
Provide an excellent browser based user interface for users as well as admins. Using the flexibility of opensource we can provide excellent tools to users like mailing lists etc. Integration between all communication media of the solution in the same screen/UI - a la gmail+gtalk but better. This will mean microphone plug-in for firefox. We aim to make everything browser based with an optional client program. We should never need a command line for anything. All admin tasks should be accomplished using a browser. Easy backup/restore/replication functionality.
Groupware integration - Am not really a big fan of this. Existing groupware seem to be too bloated. We are considering eGroupware as an optional plugin but not very seriously at this stage.

Our tentative concrete solution for the above concept:

Mail server: Postfix (To decide on IMAP server)
Integration with Fedora Directory server and MS-Active Directory Server
On-the fly account generation for all services - i.e. Once you create an account in the directory server, as soon as the user logs into any of the services after he is given a new Directory login, he should be able to use any of the services as required without the admin having to explicitly create mailboxes etc. The mail server/IM server have to be aware that a new user has logged whose 'mailbox' and/or other directories do not exist and create them automatically. Of course the mail server itself need not have this capability but the product will have this through add-on scripts or some such mechanism.
Jabber or chatty of eGroupware? for IM.
VOIP - asterisk. we want to give ekiga/gnome meeting like capabilities in the browser itself.
Video conferencing: H.323 based. (To be explored)
Based on either Fedora or CentOS. - Why? - Excellent community. Proven server side capabilities - Huge user base, we are comfortable with Red Hat as we have been using it since 1998. Lots of 3rd party rpms if functionality needs to be extended. Aiming for binary compatibility for ease of use/administration.
Totally standards based.
Totally GPLed - probably GPL v3 (To be decided.)
Further features to include security/encryption for all forms of communication and ability to record/store voice/video calls as well as IM sessions.

Important issues:

Aim for browser based functionality and hence will try to give VOIP/Video extension to Firefox. If this seems to cumbersome/ time taking, we might provide installable clients and later move to complete browser based interface. We are exploring voice plugin fot Firefox right now.
Completely standards based.
Modular so that the UI can be integrated into Intranets - Look at portlet based UI to the features. That way these can be integrated into other Intranet apps like Wokflow, Proj mgmt etc.
This product will be an excellent fit for thin client/terminal kind of apps. This solution should work even from 'kiosks'.

We are starting an opensource project soon on sourceforge.net for individual components like browser plugins. We will release the code for actual product too.

Where Microsoft is still winning over Linux in the Enterprise

2007-08-24T12:37:00.000+05:30

There have been occasions that I have been forced to migrate some GNU/Linux based stuff to MS products and after migrated unwillingly, I have come to understand why MS is still competitive and does indeed provide value.
It DOES provide value that Linux is just now starting to provide. Don't send me the flame mail till you are done reading the full article.
We have identified the gaps and our firm is working on the GPLed products to resolve some of the issues. We have planned about 4-5 major products and a couple of minor ones which will in turn be used in the major projects. All of them will be GPLed. Please join and contribute if you can. All help will be well appreciated.

I have been consulting on GNU/Linux based solutions for almost 7 years now and using GNU/Linux since 8-9 yrs now. I do not have to go into a lot of detail about the stability/security and flexibility of GNU/Linux. All of them are very well documented and I am simply awed by the prodigous amount of excellent quality code the communities have produced.

Having said that, there are some of the really painful issues that need to be addressed:

Simplicity for end user. Think Google. I think this is the single most important thing for the success of Google. Simplicity of use will make user WANT to use our applications.
Spectacular lack of competent/qualified sys-admins for GNU/Linux.
Dental surgery like pain in configuring and maintaining enterprise class servers - especially openLDAP, Mail servers, Firewalls. openLDAP has been by far the biggest problem to configure and maintain. It defies imagination on how we can have such a fantastic desktop like KDE but fail so miserably when providing a simple way to manage openLDAP. (Don't jump on me that Linux is simple or that I am a moron. I may well be a moron - don't bother pointing it out repeatedly.)
Lack of application integration. This is a fallout of the excellent choice available. We simply cannot integrate all the Free apps. This is not a 'crib'. I shall discuss what R-Knowsys Technologies will be undertaking to solve some of these issues. We need to encourage choice but also ensure that the end user(admins) are not very stretched. And this is where there is excellent opportunity for entrapreneurs.

Of the ones listed above I think Directory Services will take the first place. The number ONE reason where we face problems with large scale Linux deployments in Enterprises is Microsoft Active Directory and its integration with the rest of the services and most importantly ability to manage hundreds of desktops with the help of Group Policies.
And this is where Microsoft provides real value.

Before we jump off the deep end and start plugging a particular technology, let us try and explore the domain.
What are the really essential pieces in a corporate IT environment? The answer for this can be as simple or as complex as you want. The components can vary widely based on the following aspects of the enterprise:

Size - In terms of turnover/people.
Type - Non-Profit/for profit.
Indusrty/sector - requirements for manufacturing will be very different from requirements of a pharma co.
Budget - Is IT essential, how important/relevant it is to the enterprise.
Collaboration/communication needs of the employees. What kinds of communication.
Mobility of employees.
Security.
Statutory requirements/audits.
Employee profile whether they are IT savy....
Specific applications which run only on Windows and do not have equivalents for Linux.

More factors may affect decision but we may assume just the above for simplicity.

Now obviously we cannot come up with a fixed list of requirements which can be
used for any organisation.
Let us look at some basic requirements for a reasonably modern enterprise with 15 desktops.

Desktops, Office suite, Password security.
Internal network (LAN, WAN).
Basic networked applications - E-Mail, Intranet apps(Browser based).

Let us figure out the administration tasks associated with just the basic requirements:

Ensuring that desktops are up-to-date with patches.
Configured with the correct time.
Making sure that the hard disks are scanned from time to time.
Protection against virus/worm/malicious threats.
Data backup.
Access controls.
Scheduled maintenance for OS/Apps. Cleaning corrupted apps/files.
Ensuring that the user cannot fiddle with settings that he is not expected to fiddle with: IP addresses, proxies, add/remove applications etc.

Let us assume for illustration that the firm is a biotech firm.
Assuming that I am a sysadmin for these people, I might look at any popular desktop Linux distro like Ubuntu/Suse and be reasonably free from headaches.

Now let us change the scenario to more like what we face regularly. The number of employees > 500 and more often than not 1000 or more.
There needs to be a

'corporate standardised desktop with the mission statement diplayed prominently'
Since the firm deals with research work/sensitive IP/secrets/Govt/defense related work, email policies are such that

only some employees can receive emails from external sources.
Others can send and recieve email internally only.
Of these, only some can send attachments and that too only internally.
People in marketing can send attachements outside but size not greater than 500KB and that too only pdf docs.

Only top mgmt. can recieve attachments. Top mgmt can send attachments above 2MB.
Some employees are travelling all the time and need to access their email securely over the net.
There are about 20 offices worldwide and mgmt. travels all over the place and need to log into the corporate network from any office and work seamlessly i.e. they should not have to configure proxies/mail servers, IP address or ANYTHING.
Now Internet access is restricted based on login and it is restricted to need, seniority and even the actual sites allowed to each person. Internet access needs to be throttled based on user and application. No free email sites should be allowed to a large section of the employees.
Users should not be able to transfer files outside the corporation.
There are about 30 Intranet applications which need to be integrated into a centrally authenticated and centrally administered Intranet.
New users should be automatically given access to all applications (as per privileges) the day they join the organisation.

Do you really think Linux is up to this? Always keep in mind that the sysadmins are not kernel hackers. They cannot be expected to know 'C' or even advanced shell scripting. They just have basic scripting knowledge.

The best part is that I am not making this up. We did face this task from one of our customers. The customer was really pushing for MS products and though we resisted in the beginning, we saw the light and deployed MS-Server-2003 and standardised on Windows XP SP2 on all desktops.

We tried to retain their long serving Qmail server which had run for 2 years wihtout a reboot(I checked the uptime).
Their current IT Head wanted Exchange, we tried to save them money by continuing with Qmail on Linux. We went nuts trying to implement their email access requirements in Qmail. Even if we had implemented the requirements in Qmail, they couldn't get a decent sysadmin for maintaining Qmail. We were not into the sysadmin space either. In the end we gave up and implemented Exchange.

Before the project,

No of servers --- 2Nos, P-4 servers with 512 MB-RAM each. And 1 P-3 server with 512MB RAM(yes in 2006)
Server-1 --> Qmail with about 1000+ email accounts + Spam control + Default gateway for 100 users (squid + iptables firewall with NATing).
Server-2 --> Default gateway for 900+ users (Squid + firewall including NATing).
Server-3 --> LDAP(addressbook) + ftp server + internal gateway and routing. (P3-512MB)

After the migration, their ADS server alone was a dual opteron RAID-5, 8GB RAM box. and they had ADS replication servers in all locations accross the country.
Their hardware acquisition alone ran into tens of thousands of dollars and you can guess the licensing costs for Windows for 1000+ users -

Desktop
Office suite
Exchange
Spam control software licenses
Antivirus (Both server and desktop)
Client access licenses for email - this was unbelievable I couldn't believe that AFTER they had purchased licenses for both Exchange AND outlook, they still had to fork out for CAL.
Sharepoint + Groove.
list went on.......

Now the organisation runs pretty much completely MS-products from desktop to servers and R-KNowsys was instrumental in migrating all of their infrastructure.

From:

Desktops: Win98/ME(Desktop)
Servers: WindowsNT (Authentication), Linux servers(Mail/squid/LDAP-addressbook)

Desktop: Standardised WinXP(desktop),
Servers: Windows 2003 server + Active Directory.
Firewall - Hardware based. ISA failed miserably here even with 8GB RAM and 2 dual core Opteron chips.

Was it a loss for Linux? Definitely.
What could Linux have provided for a win? What is R-Knowsys Technologies doing to improve the situation? Where is the entrepreneurial opportunity?
How can we ensure that Linux provides unquestionable value to enterprises? More on this in my next blogs.

What ails GNU/Linux deployments in the Enterprise

2007-08-10T13:42:00.000+05:30

Date: 21-Aug-07

For the record, I am one of the founders of a small IT company providing GNU/Linux based solutions for small and medium sized enterprises. (Also a couple of huge enterprises.) We do almost exclusively GNU/Linux - Ruby On Rails (RoR), MySql, Postgres, JAVA/J2EE/Eclipse, Apache/Tomcat, JBoss and in some cases Oracle on GNU/Linux. We also undertake some short term system administration work when requested by existing clients.
We primarily work with Fedora and have recently started working with Suse and Ubuntu.
Before founding our startup we were employed in large IT services firms and have worked in India (Bangalore, Bombay, Gurgaon, Noida), USA(Bay Area), and UK.

This blog is a record of my entrepreneurship as well as thoughts on how GNU/Linux can be made more relevant to end users and how we have been successfully making a living almost exclusively with Free Software. I believe that there is a huge opportunity for entrepreneurship opened up by the Free software movement.
I will mention some real life case studies where we saved tens of thousands of dollars for our clients using free software and cases where free software couldn't compete with proprietary solutions even when the cost of the proprietary solution was 10-15 times more and ran into hundreds of thousands of dollars. We decided to build some free software components just because we were frustrated with the existing free software alternatives.
Most of my observations are based on our experience with deploying GNU/Linux in various non IT and IT enterprises. For some of these business, IT was essential, for others IT is not essential at all. For some, IT was their primary source of income.

We are and have been very actively evangilising GNU/Linux based solutions to our customer but after a couple of tough deployments, we have grudgingly come to accept that proprietary solutions provide excellent value in quite a few cases - even considering high cost. Now in our solutions, we do recommend proprietary components over free ones - especially MS-ADS. (Flame on.....)
When we evaluate solutions for a customer, we look at -

'What are the ones that work well for my customer and are they painless and risk free?'
If that is satisfied, we look at - 'How much does it cost ---- cost vs. benefit + flexibility', 'How easy is it to maintain/upgrade etc - the much famed TCO.' 'HOW EASY IS IT TO GET THIS APPLICATION ADMINISTERED?' This is pretty important and I/we have seen quite a few deployments of GNU/Linux rolled back due to this factor alone. More on this as we go along.
After that - 'How does it play in the ecosystem - interoperability'
Then comes 'freedom' - Why does it come last? More often than not freedom is not even relevant. Freedom becomes relevant only at the level of data storage formats. Let me illustrate - If tomorrow MS-Exchange or MS-Word stored data in open formats, I wouldn't be too concerned about what the program did (unless the project was for Govt./defense/public security purpose). I will not even take a 'Joe User' example to illustrate this. I will take a 'Joe programmer' example. Why will an average coder be concerned whether his/her development editor is proprietary or free as long as its output is plain text format? Eg. - Notepad vs. Gvim. (Assuming Notepad is proprietary) Now extend this example to any computer user in any industry - say an assistant in a law firm, a manager in a corporation or a teacher in a primary school.

Freedom to tamper with the source code? - That is the last thing on the mind of any IT infrastructure manager in a small to mid level IT co. I will have to pay for customising the product in one way or the other - either to the s/w co., to in-house developers or contractors. In a short/medium term business perspective, it is pretty irrelevant to me.
(Of course this thinking may lead to monopoly and more bad things but assuming that I am a average IT/MIS head in a small/medium/huge corporation, the consideration that my vendor may be trying to monopolise will be the least of my worries if it works most of the time.)

The 'Freedom' concept applied to software becomes relevant only when formats are closed and in boundary conditions.

Contrary to popular opinion, I do not think that the server war has been almost won by GNU/Linux or that MS is having trouble selling servers to Enterprises. Linux is not even a threat in quite a few areas. Linux is appreciated in ISPs and Datacenters and as a host server for deployments of Oracle/Apache. It may be making progress but does not have some really essential pieces required by servers in an enterprise network. These features were there in Novell Netware almost a decade and half ago and some features started appearing in Windows servers since 2000.

The biggest advantage MS has in the enterprise is its Active Directory Server. It is really a boon for a understaffed IT/MIS department to manage hundreds of desktops and integrating a range of networked applications. LDAP fails terribly not because of any lack of features but because of the absolute absence of a excellent interface to manage it. There is a very urgent requirement for an free directory services with a half way decent UI which is tailored to the regular everyday requirements of sysadmins. The few php based ones I have seen are not good enough. The UI should try more and more to remove me from thinking about implementation details - like a good API. For a power administrator, we have the text configuration files anyway.
ADS is being leveraged to push the rest of its enterprise products.
The other big advantage MS has and is leveraging like crazy is MS-Sharepoint. To some extent Groove too.
If we can make excellent Free software products which can compete effectively against these two, I know we can hope to compete against MS productline as a whole. If there are no free software products to challenge ADS and Sharepoint, MS will always be the leader and Free software will always remain a good cost saving second choice for most applications.

In my opinion, Free software lacks a couple of very important aspects -

Focussed interface of the community with end-users(a pretty decent job being done by Redhat, Ubuntu and hopefully by Suse) and genuine service where we are listening to the customers issues and solving them instead of just giving the customers the 'freedom to hack the source'. Most people will not have the resources, ability or inclination to handle source code.
Eco-System of a chain of ISVs, service providers and trained people. Trained people is really the biggest bottleneck for free software adoption. Most of the other issues are being resolved and some have been resolved. Unless there is a very focussed attempt at churning out well trained 'free software deployment and administration personnel', free software uptake will be very slow and may be a sitting target while this is in being built up. I had an exchange of email with RMS back in 1998 but I was too young to take it up myself.

Relevancy to a user is NOT JUST about 'cost' and 'freedom'. Though 'cost' and 'freedom' are important, cost comes into the picture only AFTER the utility of the solution is demonstrated and it is comparibly simple to use. Freedom comes even later than that.

In the coming posts, I will explore what specific issues are essential to making a successful Linux based deployment in enterprises. We will look at (in my opinion) what pieces are missing, incomplete, under-leveraged.
What are the biggest strengths of current proprietary solutions(specifically Microsoft, Oracle, SAP).
What is needed for free software applications/solutions to provide clear value to users based solely on utility and technical merit without considering cost at all?
What are our plans at our firm for building those kind of solutions.

Next blog: The essential pieces of Enterprise IT infrastructure - Where current Free Software is very good and where Free software still doesn't make the grade.