Today, a new open source company launched, SourceLabs, which promises to deliver “dependable open source systems.” According to their Web site, they are focused on testing, certifying and supporting open source systems. The company is founded by BEA alums, Byron Sebastian, Cornelius Willis and Will Pugh, and has received $3.5MM in funding from two VCs.

SourceLabs is taking a new approach to open source by creating services and solutions that allow companies to depend on any open source projects, irrespective of the brand, author or configuration. SourceLabs states that they will be creating common configurations of popular open source products along with providing sophisticated support options to customers.

To differentiate themselves, they have formulated six promises to their customers and the open source community.

• We promise to not lock you in, ever
• We promise to keep our company open
• We promise to keep our egos separate from technology
• We promise to provide only software that we've tested and certified
• We promise to be accountable
• We promise to make your software work

In a TSS discussion with the founders, we asked them about the technologies that they were going to initially focus on. While they did not announce specific technologies supported, they indicated a focus on server infrastructure and applications, so that might encompass a fairly broad opportunity, including databases, containers, scripting, and other application server technologies.

Here is their launch press release: http://www.sourcelabs.com/SLPR0104.pdf

Here is their new Web site: http://www.SourceLabs.com

Editor’s Note: This could be a company that is set up to directly compete with JBoss, Inc. and Red Hat. While these companies have invested significant energies into developing brands and a small set of brand-supported products that they endorse, their view point and scope of the world of open source is limited. Will their model and open approach to the market take off? What does the community think?

Finally no egos.
So are you going to have machines do everything?

"No Ego" means that we will support software that customers and the marketplace choose, rather than software that we write. This means we won't run into issues of spreading FUD to promote one technology or another, or having a "Not Made In Here" additude. We want good technology to decide which products win rather than good marketing.

We are looking to put all our engineering efforts in making the best support processes, and system testing tools as humanly possible. But, it will be humans that write this code :)


    Will Pugh
    Chief Architect, SourceLabs

It's a good idea and it'll definitely be nice to have fewer egos.

The only potential pitfall I see with this (from a technical perspective) is that the people who write the code also happen to be the ones who can best support it.

I find it very hard to believe that anyone else could sell JBoss support as well as JBoss can. Large companies who want real peace of mind will want the people who wrote the product supporting their live 24/7 production server.

One way this business could work is if it ends up being an aggregrator of support. So instead of buying support from 10 different open source projects, you can buy support from one place, and SourceLabs would in turn buy "wholesale" from the other open source projects. That way, SourceLabs can filter the 60-80% they can know (and is documented) and they can go to their wholesale for the other 20-40% no one else knows besides the developers.

Good luck!

And if you guys ever want to look at a J2EE enterprise portal that already offers professional support and may fit into the latter business model, check out Liferay (http://www.liferay.com).

Brian Chan
Chief Software Architect
Liferay, LLC

I find it very hard to believe that anyone else could sell JBoss support as well as JBoss can. Large companies who want real peace of mind will want the people who wrote the product supporting their live 24/7 production server.

JBoss is an exception, however.

For example, these folks could support Tomcat, or any of the myriad of OSS projects that are becoming mainstream.

What they're providing as a service is simply known contact point for the software they support. With mature products, a lot of support is configuration or documentation based rather than actual code changes.

The risk is that they actually do find a "bug" and actually fix it and release a new version to their client, but then the parent OSS project decides they don't want to incorporate the patch, or do it in a different way that may well "Fix the bug", but does so in a way that is incompatible with the fix they sent to the client. At that point they either need to fork the project or convince the client to upgrade.

So, as with many things, it's all about expectations of the clients.

Idea is good and viable

Why it would work
1)Customer would get "one place" to integrate,get solution using Open source applications.
2)Customer would prefer Open source as issues can be addressed and resolved.
3)Faster adapation of open source.

Why it wont work
1)Lack of approval from the open source vendor like JBoss/Tomcat.
2)Higher Service cost(depends how much they gonna charge).


Santosh Panda
UPCO,www.upco.co.uk

Why it wont work
1)Lack of approval from the open source vendor like JBoss/Tomcat.
2)Higher Service cost(depends how much they gonna charge).

JBoss is an open source project under a Free Software Foundation license. Anyone can build a business around it, regardless of someone else's approval. On the other hand, most companies would rather go to the source, which is JBoss, Inc.

Tomcat is Apache licensed. You don't need anyone's approval for that, because there is no "anyone". A good percentage of the app servers out there use Tomcat for their servlet engine.

The business question is whether there is money to be made. The JBoss, Inc. approach seems to show that there is .. training and support are pretty high margin products. On the other hand, JBoss works to maintain its relevancy by investing in those projects .. if this new company just does support, I'm not sure if it will be able to get the same level of traction.

Peace,

Cameron Purdy
Tangosol, Inc.
Coherence: Shared Memories for J2EE Clusters

Hey I noticed some of you guys worked for BEA Marketing. Were any of you around 2 years ago when BEA sent FUD Marketing about JBoss, claiming open source could endanger your intellectual property? I remember that, how about you?

I have lost a lot of passion about this since it seems inevitable, but the fact that 4-5 of these guys have decided that working for a "services" company is better than working for a "product" company is a sign that intellectual property in software today is losing value instead of gaining value.

Obviously, there is a natural trend toward commoditization. However, without any controls (like patents) on the rate at which the shift occurs, commoditization will quickly lead to an industry dominated by services companies that lock companies into low-tech, services-intensive solutions rather than the proprietary hardware, databases and app servers of the past.

Matt

I don't think software IP is losing value (just look at the ongoing speculative bounty-hunting on land-grab software patents). I am concerned that an upwardly mobile pyramid of large companies standing on the shoulders of open-source giants will threaten the diversity and viability of open-source software. See here for a further exploration on this important issue.

hhmm.., i lov the idea.. often v see hw big corps lock innocent customers into some proprietary technologies and keep their market share.. most of the companies manage their market thru this strategy(??!!) only :-)
  nice to see someone saying they wont lock the clients into something.. atleast in their mission statement.. a healthy trend indeed.

Contrary to some others in the forum, I think it is a great idea and long overdue. Given that a number of open source tools, products, and frameworks have been mainstream for a while, it is surprising that no one had thought of it before.

IMO, they need to operate at a level above JBoss/Redhat - professional open source at a higher level encompassing a wider range of products. The focus should be more on consulting than bug fixing/support - at least in terms of interfacing with clients (needs to be backed by thorough evaluation and testing of the products). I am sure there are a lot of projects looking for the appropriate and reliable open source tools and don't want to spend a lot of time evaluating and testing the hundreds that are available.

Given that BEA alumni (some with marketing background) have started it seems to indicate sufficient due diligence and knowledge of market trend. Could be the direction the industry is taking. It is certainly great for open source and for for those who want to use open source and minimize associated risks.

Kalyan

Core Developers Network - Dont they do the same thing?