The Jeff Pulver Blog: Guest Blogger: Kfir Pravda - "To Standard or not to Standard?"

« Fall 2006 Video on the Net Featured Speaker: Jason Krikorian, co-founder, Sling Media | Main | VON Enterprise: December 4th-6th in Atlanta, GA »

December 01, 2006

Guest Blogger: Kfir Pravda - "To Standard or not to Standard?"

So you gathered a bunch of telecom freaks, rented a basement, and saved some budget for cold Pizza. You are going to conquer the world with your amazing application that changes the way people consume media and communicate - forever. Chambers is going to beg you for a job, and the guys with the funny name from Estonia will have wished they stayed in P2P file sharing applications when you're done.

Now is the time to get down and dirty with the little details - such as - are you trying to build a whole new ecosystem, or ride on the waves of others?

More specifically - are you going to create your own proprietary protocols, or base your product on open standards?

One of the biggest mistakes is to think that this is a technical question that an engineer should answer. The truth is that this question is mainly a business and strategic one. It pretty much depends on the way you see your future - do you want to be an ant in the grass, with a chance to become the next big thing that captures the market? Or would you rather ride on the back of the elephant, with a chance to play a major part in an industry created by others (with deeper pockets)?

I have to say that there are a lot of pros in going standard. First of all, you can reduce your development time by using the accumulated knowledge of the industry. The knowledge you can tap when working in a standard environment will always exceed any amount of engineers and technology experts you can possibly hire.

Second, in case your application is based on a Network Effect, like most of the communication products, you can rely on the marketing dollars of others to educate the market. Then, you just need to find a niche where you gain cash and exposure (in a way, the "crossing the chasm" concept).

Third, you might be able to shorten the time to exit. If you base your products on standards, a company which is interested in buying you will have a much easier life in integrating your products in their organization and product line (based on the assumption it also works on standard based products).

Well, this would have been a great post if those annoying guys from Skype didn't come with their amazing application. You see - they did it all on their own, and at the end of the day - made my mother use VoIP - before any other SIP based product. They focused on user experience, and still managed to beat the rest of the VoIP techies to the desktop.

If so, maybe the standard world isn't that great? First, it takes ages to draft standards. Then, the standard bodies are dominated by the big players, which make the life of the little guys harder - as they have different agendas then helping a young start-up to rise. And last but not least, it is not trivial to find a niche in a standard based industry, especially for a small company. When standards reduce technical competitive advantage, marketing dollars kicks in - an area in which a small company will usually loose to the big guys.

So, here is the question: If you would develop a new video conferencing application, the next VoIP system, or any other communication related product - what will be your choice? To Standard or Not To Standard?

---

Kfir Pravda is an independent consultant, who facilitates cooperation between media and technology companies. On top of his day job, he heads the marketing group of IMTC (www.imtc.org), and loves crime movies and whisky. He can be reached at: kpravda at gmail.com.

Tags: IMTC, Kfir Pravda, standards, Jeff Pulver

Share this post:

Posted by jeff on December 1, 2006 05:44 AM | Permalink

Additional resources: Watch PrimeTime TV Shows | Watch the Jeff Pulver Show | Jeff's Qik Videos

Comments

The standards question is a red herring when applied to Skype. Standards encourage mass adoption by allowing an ecosystem of interested vendors to develop a market in tandem. No doubt that Skype did a good job with their application, and created a great user experience. But they were able to succeed in a non-standard only because they had an alternate path to get mass adoption - through the installed base of Kazaa users. Kudos to them for being able to pull it off.

Posted by: ehernaez at December 14, 2006 04:22 AM

From Israel - I established a New Company developing an Engine / System enabling to Identify and then Detect and Billing Non-Standard and Standard Real Time Media Packets based on its Network's / Signalig Behavior (Identify real time non-standard transactions / packets such as Skype, Jajah and alike) - I am looking for strategic cooperation for both co-engineering and Co-Marketing - Open API for any Firewall, IP-Switches etc. - you can reach me at EldadCa@Gmail.com

Posted by: Eldad at December 10, 2006 06:51 AM

A thought provoking article! Standards provide the "lingua franca" to be "interoperable" and make it as a "mass" product.However, Standards also deprive the growth of "creativity". The proposals in the standardization committes are not about how good the technology is or not about how beneficial it is for the end user. Most of the times its about business gains and patents.

If the idea that you have, has compelling reasons to be profilerated into the community for example ...skype) then its great to be on "proprietary" path rather than in "standardised" path.

To summarise, its all about whether you have the entreprenural "Belief" within you (Proprietary) or with others (Standards)!

Posted by: P.S.Sanjeev Kumar at December 5, 2006 06:09 AM

Communication always reqires some sort and level of standards, whether de-facto or de-jure. And of course depending on your communication and application environment you may distribute the application (incl. some of the standards needed) over the the telecom channels (what Skype does). But the principle is not new, only that it became more practical, than in the past. We did the same thing twenty years ago, e.g. within the Videotex standard.

One remark: Any sort of generalization (such as "the standard bodies are dominated by the big players, which make the life of the little guys harder" or "standards take too long") is dangerous, as they are many-many examples also for the opposite.

Posted by: Istvan Sebestyen at December 5, 2006 05:14 AM

Kfir, nice post... Good question !

While being an avid supporter of Open Standards and Open Source, I believe there's no "one true" answer to this, and it also depends on whether it's an "infrastructural" (like Network elements) or "End-user" product.

When you play the proprietary game, it's all about what's *IN* your product... You are then confined to your tiny world, and are condemned to build it all.
[Obviously, if you are lucky enough to control the market, your users will end up with silly paperclips dancing on their screens when they write a memo...]

But, when you play in the "Open" field, it's a whole new game... Less of what's *IN* your product, and infinitely more to what's *OUT* of it, or actually, *HOW* you play in your universe.

Ultimately, universes provide infinite opportunities for players to fill-up empty spaces, and while doing so, create new needs and spaces to be filled...

As we talk about "HOW" you play, I'd say the real question should be "To Interoperate or Not To Interoperate ?", as being standards-based seems to be an implicit issue of this.

Posted by: Pablo 'merKur' Kohan at December 4, 2006 07:36 PM

Skype did two things:

1. Build a super friendly application and this has nothing to do with the standard
2. Proprietary protocol that enabled them to get to the market very fast

The combination of the above made them what they are or in other words there are a few dimensions to this question and the result needs to be analyzed case by case

Posted by: Noam Zakai at December 4, 2006 03:15 PM

Well as you specified, this is not an easy decision,
When you go standard you have to find the niche - this is not an easy task, and if you don't find it, you have nothing special to offer.
Still I would say - go standard, but take advantage of standards "holes" (and usually there are a lot of them) and add you proprietary stuff which will still interop with other standard implementations.

Posted by: Amir Wolf at December 4, 2006 01:51 PM

User experience comes first. Higher gains are definitely what it's all about, this is a business environment, not a club for telco engineers.
Standards & bridges can be sorted out later. Everyone knows Skype & iPod & iMode. Who (normal joe public) knows Gizmo project?

Regarding the push of the industry - it's not worth much aside from bearer technologies. For services standards take too long and you have to rely on your competitors and then you'll be eaten by an open source implementation. Internet model (get big quick) still wins.

Posted by: David Goulden at December 4, 2006 05:02 AM

Well, it is not exactly true Peter. No SIP/H323/XMPP application can call Skype users without the usage of 3rd party applications... So it is right you rely on standards, but only in things that does not affect your business....

Posted by: Kfir Pravda at December 4, 2006 04:15 AM

Good post! A tricky issue.
Skype does, of course, rely standards too. PSTN-standards, IP-standards, firewall-standards etc.

Posted by: Peter at December 4, 2006 03:46 AM

Definitely go standard. Proprietary might mean higher gains, but at a higher risk (a lot higher). Being in the standards market for most of the last 10 years, and playing around on both sides, having a standard means the work is more complex, but the push of the industry enables you to gain something if you don't do too many business mistakes.

Posted by: Tsahi Levent-Levi at December 4, 2006 02:04 AM