The vTiger fork was not for technical reasons, which is one of the strengths of open source software, but rather a business strategy, in effect “leeching” of the hype created by SugarCRM and the work they have done. SugarCRM then added an attribution clause (SPL v1.1.3) to prevent this from happening again in the future. During the mudslinging that ensued SugarCRM were criticised for the way they handled the situation and not for license changes, as it was understandable why they were implemented.

At that stage we were busy evaluating the various open source licenses to release the TenderSystem code base, and saw the genius behind this model, especially for a web user interface application, that created a win-win scenario for developers and the community at large.

When TenderSystem was released under the TPL it was not submitted to OSI for certification, as there is pressure to decrease the number of approved licenses, even though it adheres to the definitions of open source. OSI will also most probably not certify project specific licenses, as the MPL states that the words “Mozilla” and “Netscape” may not be used for a derivative licenses, thereby creating a new license even though no other wording is altered.

Since then many projects were launched under this license and some existing projects even changed to this license due to the good fit for web based applications, as is the GPL for embedded applications. Lately some critics have come to the fore saying that the license is a form of badgeware and that it encumbers the spirit of open source.

Fortunately Mitch Radcliffe from Socialtext, a projects that also release under an adapted license, has submitted a General Attribution Provision (GAP), to be attached to OSI approved licenses that make provision for it, for approval to to the OSI board. The general provision is a very clever idea, as it will not affect the OSI license proliferation problems, and hopefully a formal response of their views will be released shortly.

If this is approved it will allow these projects to embrace other licenses, such as the GPL or LGPL, will most probably be TenderSystem’s choice, instead of just the MPL as is the current scenario.

Nicholas Goodman commented on the GAP provision on his blog to also provide limited Trademark and logo rights to prevent projects from circumventing the license through Trademark usage. The problem that this might cause is that a Trademark has to be protected to remain valid, which was the cause behind the Debian/Firefox debacle, and might be better to insert a complete Trademark Usage Policy, or version, in the license.

Rick Moen blogged wrote an article at Linux Gazette that members of the OSI mailing list found an obstacle with OSI’s definition #10, namely “License Must Be Technology-Neutral”, in that sometimes a user interface is not present, such as when running a daemon. This should not be a problem by rewording the GAP to a “reasonable attribution”, the term coined in Nicholas Goodman’s blog, in that adherence to attribution is only required when a user interface is present.

Which ever way it turns out I look forward to the outcome of the process and sure that there are going to be some interesting comments to this post.

[UPDATE: Socialtext have withdrawn the GAP for OSI board revision and will submit a new license in the future.]

My personal experience has been quite the opposite, where funds nearly bankrupted ValueCard before we even got started. In our early days we had more than sufficient capital, lets call it our dot-com days. This was private funds, so no ludicrous war chest, but more than enough to go around.

The problem with having sufficient cash in the bank is that there is no initial urgency to get clients on-board as you first want to “lay a solid foundation”. So we went out and got a fancy office, kitted it out with nice furniture and started working on the business plan. As if that was not enough, a very good secretary was employed, to handle all those calls that we would be receiving from potential clients; wanting to come for a quick coffee at our nifty offices while discussing the business that they would send our way.

Because there was no urgency it took us much longer to develop TenderSystem and exceeded the budget dramatically. We also didn’t have any clients, as it was difficult to sell a vision without a product to back it up.

We then woke up one morning and realised that we were technically insolvent and still burning cash as if it was going out of fashion on our nice to haves. Fortunately we were able to salvage the situation by cutting all unnecessary costs, even though it was not so easy to renegotiate those agreements, and also got hold of some bridging finance. We were also able to get clients to use our products and services, as it is very easy to when you are “hungry”.

The end result is that the company survived, and I think became much stronger from the experience. We do not have venture capitals breathing down our necks to deliver immediate results, to the detriment of sustainable growth, and can concentrate on the company’s future.

I understand that the lack of funding normally results in bankruptcy but think that too much initial funding is just as detrimental, even though it just prolongs the inevitable much longer.

The benefits achieved by implementing a procurement system must be viewed from two perspectives, namely measurable and un-measurable savings, which some would call hard and soft savings.

Measurable benefits are the direct price savings, achieved through leveraging the age-old principle of supply and demand and by increasing the purchasing power by channeling all purchases through a single channel. These price savings are tremendous and we normally achieve an audited saving of 20% for short-term insurance replacements. There may be an additional administrative cost involved, if prices are not present beforehand and have to be received manually, but most systems will display the spread difference between the cheapest and most expensive supplier that can be used as a guide.

Secondly un-measurable benefits are the additional savings achieved due to a more efficient process. This includes decreases in human resource, management, administrative and telephonic expenses as well as a decrease in incorrect and unnecessary purchases and fraud.

Incorrect purchases are expensive to rectify as there is normally additional administration and delivery costs involved and a cancellation fee can also be levied. These purchases are usually picked up and eliminated during the approval process or when escalated to the agents administrator, when a pre-approved mandate was breached.

The possibility of fraud always exists, and does occur, in procurement departments. This can be either fraudulent personal purchases or collusion fraud where kickbacks are paid to the procurement agent. Kickbacks are very tricky to control, as they come in many guises such as a trip to the Bahamas or massive discount on items. At the end of the day it is the purchasing company that pays for this indirectly through higher prices.

I am sure that there are many additional benefits that I am not aware of, as my experience was gained with insurance replacements, so please comment as I would also want to hear your point of view.

We use TenderSystem in-house and offer the procurement solution to insurance companies on a software as a service (SAAS) basis. If there is a bug or inefficiency it does not just inconvenience our clients, but costs us time and money, so you can bet your last dollar that we try to keep bugs to a minimum and fix them as and when they are discovered.

As the project sponsor this filters down into the open source code and our community members end up with tried and tested technology. TenderSystem benefits out of the open source process as well due to bug reports and additional functionality requests, that we could never have thought of on our own.

This win-win scenario is not present in proprietary software were developers do not feel the pain experienced by their clients. Developers are usually shielded from clients, so some critical problems, that might seem irrelevant to developers, are never prioritized and therefore never fixed. Additional features are also only added if there is a benefit to the development company, such as a barrier to entry for competitors or additional potential revenue, as there are initial development and future management costs involved.

It becomes even worse when a developer and their client’s best interests are diverged, as was the case with Internet Explorer. Microsoft saw, and most probably still see, the Internet as a treat to their monopoly on the desktop operating system market. Once the competition was eradicated, all development was shelved on the most prominent web browser and it took an open source project called Mozilla Firefox to put pressure on their dominance.

Open source users also benefit by supplying back bug fixes, feature contributions and improvements as it will be included, and tested, in future versions without having to be patched manually.

This results in more user friendly software produced through the open source process.

TenderSystem blog will be a place for us to discuss TenderSystem, procurement solutions, open source and software development. This will not be an extremely active blog as there are not any set periods to write, but rather a place to think tank new ideas.

If no new ideas are born from this blog I will scrap it quickly as there are already numerous other blogs on the Internet to keep us occupied. Please comment and give feedback on the posts so that we can all gain and get insight together.
Should you want to publish your own blog please let me know and I will arrange you with the necessary access.