Comments for Fusio9 Blog http://www.fusio9.com/blog Fusio9 Blog: Explorations of a serial technology entrepreneur Fri, 21 Nov 2008 17:40:25 +0000 http://wordpress.org/?v=2.5 Comment on Effortless scalability by Emile Cambry http://www.fusio9.com/blog/2008/04/16/effortless-scalability/#comment-9 Emile Cambry Thu, 17 Apr 2008 07:14:05 +0000 http://www.fusio9.com/blog/?p=12#comment-9 should be "more value" not "move value" should be “more value” not “move value”

]]> Comment on Effortless scalability by Emile Cambry http://www.fusio9.com/blog/2008/04/16/effortless-scalability/#comment-8 Emile Cambry Thu, 17 Apr 2008 07:13:42 +0000 http://www.fusio9.com/blog/?p=12#comment-8 Hey Joe. It's great that you posted on this. I actually talked a little about this yesterday on my blog. I really think that this movement is groundbreaking and the move value these bigcos provide, the more the Joe Dwyer's of the world can focus on solving real problems, and furthering innovation. Hey Joe. It’s great that you posted on this. I actually talked a little about this yesterday on my blog. I really think that this movement is groundbreaking and the move value these bigcos provide, the more the Joe Dwyer’s of the world can focus on solving real problems, and furthering innovation.

]]> Comment on Effortless scalability by admin http://www.fusio9.com/blog/2008/04/16/effortless-scalability/#comment-7 admin Thu, 17 Apr 2008 05:37:59 +0000 http://www.fusio9.com/blog/?p=12#comment-7 J - I very much agree that scalability is not free, and tradeoffs have to be made. Then again, many web applications have similar needs, and therefore (within reason) can probably tolerate some generic choices. Similarly, Java works for most developers needs, and if not they can choose a language with fewer abstractions. But more importantly, your comment brings up an excellent point... assuming the market does head toward effortless scalability, there will probably be several types of cloud systems. For example, financial transactions have specific requirements such as low latency, and therefore I can see a sort of "financial scaling cloud" in the market. Meanwhile, social media sites have different needs such as robust scaling for media delivery, and might have a different scaling cloud. I think a few types of clouds might cover enough of the market. Of course, there would always be some applications that require fully custom development for one reason or another. J -

I very much agree that scalability is not free, and tradeoffs have to be made. Then again, many web applications have similar needs, and therefore (within reason) can probably tolerate some generic choices. Similarly, Java works for most developers needs, and if not they can choose a language with fewer abstractions.

But more importantly, your comment brings up an excellent point… assuming the market does head toward effortless scalability, there will probably be several types of cloud systems. For example, financial transactions have specific requirements such as low latency, and therefore I can see a sort of “financial scaling cloud” in the market. Meanwhile, social media sites have different needs such as robust scaling for media delivery, and might have a different scaling cloud. I think a few types of clouds might cover enough of the market. Of course, there would always be some applications that require fully custom development for one reason or another.

]]> Comment on Effortless scalability by Seltzer with an H http://www.fusio9.com/blog/2008/04/16/effortless-scalability/#comment-6 Seltzer with an H Thu, 17 Apr 2008 05:24:02 +0000 http://www.fusio9.com/blog/?p=12#comment-6 One of the biggest challenges -- and you touched on it-- is the developer's ability to customize. Scalability is not "free", i.e., there are inherent tradeoffs that are made when building a system to scale. If scalability is effortless, it implies that a choice for each tradeoff was made. For many appications, this may be fine. But I can also think of a lot where this may be problematic, especially high performance applications. therefore, the ability to customize-- which I'll rephrase as an ability to control the tradeoffs that are made-- I believe will be crucial. Unfortunately, I think this makes the solution hard to architect. A smaller example is Java. Java abstracts away memory allocation and cleanup from the developer. A nice perk, but the developer forefeits control over when the garbage collector runs, which means that resources are consumed and freed on a whim at run time rather at the behest of the developer. Again, this is fine for a lot of appliations, but not for all. j One of the biggest challenges — and you touched on it– is the developer’s ability to customize. Scalability is not “free”, i.e., there are inherent tradeoffs that are made when building a system to scale. If scalability is effortless, it implies that a choice for each tradeoff was made. For many appications, this may be fine. But I can also think of a lot where this may be problematic, especially high performance applications. therefore, the ability to customize– which I’ll rephrase as an ability to control the tradeoffs that are made– I believe will be crucial. Unfortunately, I think this makes the solution hard to architect.

A smaller example is Java. Java abstracts away memory allocation and cleanup from the developer. A nice perk, but the developer forefeits control over when the garbage collector runs, which means that resources are consumed and freed on a whim at run time rather at the behest of the developer. Again, this is fine for a lot of appliations, but not for all.

j

]]> Comment on if (business strategy != subject) { echo failure; } by Seltzer with an H http://www.fusio9.com/blog/2008/04/15/if-business-strategy-object-echo-failure/#comment-5 Seltzer with an H Wed, 16 Apr 2008 15:30:45 +0000 http://www.fusio9.com/blog/?p=11#comment-5 Ethernet and Windows are the two examples I keep in my head to remind myself that lousy technology can win if it becomes a standard. In other words, network effects and positive externalities are stronger than simply better or elegant technology. Both ethernet and windows are technically inferior when compared to peers, but they are insanely financially successful. Also reminds me of the old adage, "don't make perfect the enemy of good." Ethernet and Windows are the two examples I keep in my head to remind myself that lousy technology can win if it becomes a standard. In other words, network effects and positive externalities are stronger than simply better or elegant technology. Both ethernet and windows are technically inferior when compared to peers, but they are insanely financially successful.

Also reminds me of the old adage, “don’t make perfect the enemy of good.”

]]> Comment on The social web revenue inflection point by The Coral Reef of Future Internet Innovation and Disruption « These two cents http://www.fusio9.com/blog/2008/04/13/the-social-web-revenue-inflection-point/#comment-4 The Coral Reef of Future Internet Innovation and Disruption « These two cents Wed, 16 Apr 2008 09:09:18 +0000 http://www.fusio9.com/blog/?p=8#comment-4 [...] heard the theory that in order to succeed in today’s Internet economy, you have to take a platform approach to providing web services and applications. The marginal cost of an extra unit of software or code [...] […] heard the theory that in order to succeed in today’s Internet economy, you have to take a platform approach to providing web services and applications. The marginal cost of an extra unit of software or code […]

]]> Comment on The social web revenue inflection point by admin http://www.fusio9.com/blog/2008/04/13/the-social-web-revenue-inflection-point/#comment-3 admin Tue, 15 Apr 2008 02:43:07 +0000 http://www.fusio9.com/blog/?p=8#comment-3 Good point, Emile. I think that balancing the revenue generation with continuous enhancement of the value proposition is critical. Facebook has had challenges instituting certain revenue generation initiatives, and I think this will continue to be a problem for many social networks. For an example of site that utilizes contests very effectively, check out Brickfish.com. I mentioned that site in the post, but did not point out that one of their key ingredients is contests. Good point, Emile. I think that balancing the revenue generation with continuous enhancement of the value proposition is critical. Facebook has had challenges instituting certain revenue generation initiatives, and I think this will continue to be a problem for many social networks.

For an example of site that utilizes contests very effectively, check out Brickfish.com. I mentioned that site in the post, but did not point out that one of their key ingredients is contests.

]]> Comment on The social web revenue inflection point by Emile Cambry http://www.fusio9.com/blog/2008/04/13/the-social-web-revenue-inflection-point/#comment-2 Emile Cambry Tue, 15 Apr 2008 02:26:45 +0000 http://www.fusio9.com/blog/?p=8#comment-2 Interesting post. I agree wholeheartedly with your assertion that social media sites should wait until they have some significant traction before pursuing monetization. The unfortunate consequence to users is that they are almost led on to the fact that their experience would be unencumbered and as soon as it gets fun to use, because everyone uses it, they force ads down your throat. I think what is not often studied by start-ups is really understanding where the extra value lies in their product and charge those who are seeking the out of the ordinary value, such as extra features. Also, I'm curious why web contests hasn't taken off like it was predicted? I still think that avenue is not explored to the extent that it could be and is one of those features that provide significant value that could be extracted by advertisers, and provide value for the user base. Interesting post. I agree wholeheartedly with your assertion that social media sites should wait until they have some significant traction before pursuing monetization. The unfortunate consequence to users is that they are almost led on to the fact that their experience would be unencumbered and as soon as it gets fun to use, because everyone uses it, they force ads down your throat. I think what is not often studied by start-ups is really understanding where the extra value lies in their product and charge those who are seeking the out of the ordinary value, such as extra features. Also, I’m curious why web contests hasn’t taken off like it was predicted? I still think that avenue is not explored to the extent that it could be and is one of those features that provide significant value that could be extracted by advertisers, and provide value for the user base.

]]>