3 ms·

I wrote a hypothesis [1] on the cause and effect of the cycles Chris mentions. With those assumptions about the separation of user and their data in mind, I mak

I wrote a hypothesis [1] on the cause and effect of the cycles Chris mentions. With those assumptions about the separation of user and their data in mind, I make the following near term technology predictions: - a swing back to private cloud will be kicked off by container/microkernel technologies, starting the largest cycle we've seen to date in terms of growth and value - public cloud computing growth will slow slightly and will have to refocus on lower privacy needs use-cases (or die trying) - IoT and cloud computing will start to merge as a market, where the compute resources of your IoT devices are used to operate on the data other IoT devices nearby - cryptocurrencies and the blockchain will finally find a home in securing the processes behind cloud provisioning of what was known previously as SaaS software - cryptocurrency deployments make open source project's revenue models sustainable - open source hardware, including circuitry, becomes ubiquitously available via physical printing processes, which then drives the IoT and cloud computing markets even higher. - the digital nomad lifestyle explodes as a result of the changes to the open source model - startups and the VC culture in Silicon Valley will be forced to retool their financial strategies and software business models for accessing decentralized markets - a meltdown of global financial markets will be the only thing which enables bringing positive, decentralized change to those financial markets. As John Chambers, chairman of Cicso said a few years back, "You're going to see a brutal, brutal consolidation of the IT industry". Better buckle up. [1] https://medium.com/@kordless/an-ecology-of-cloud-1cedfa326b8c https://medium.com/@kordless/an-ecology-of-cloud-1cedfa326b8... BTW, there is only one thing that will make me stop coming to HN and participating in the community, and that's when I see people's opinions on something downvoted. Downvote if you must, but keep in mind it's a game of blame if the post you are downvoting is someone's opinion or view on something, which itself is blameless.

eitally 11y ago
All of the points subsequent to your first one conflict with that conjecture, that we'll see a huge swing back toward private cloud & on-premise DCs. Speaking from where I saw the industry [when I was a part of it] and now from what kinds of CIO/CTO conversations I see the GCP AMs/SEs having, I don't think we're remotely close to a swing back. Enterprise Capex budgets move VERY slowly most of the time, and it will take a multiple years to alter current momentum, which has finally reached the point where it's almost a foregone conclusion for many industries that on-prem DCs no longer make sense. This opinion is driven almost as much by the increasingly compelling portfolio of enterprisy SaaS products like Netsuite and a plethora of workforce management, CRMs, analytics/visualization, sooooo much more of what has historically been internally developed business productivity apps.
- kordless 11y ago
  There are a lot of people who think public cloud is the bee's knees and is going to continue to grow without bounds. The fact is, private cloud revenue is 30X as big as public cloud. It's this big because of previous growth cycles. Also, there's the consumer market to consider, just like in previous cycles. Public cloud is simply multi tenant computing on data managed by someone else away from where the customer created it. I'm predicting data will come home to roost on single tenant hardware (which comes in by way of IoT) and the code that has been called SaaS will follow it there. We'll still have managed services, they'll just run near us, instead of on compute owned by someone else. Regardless of whether you agree with that or not, it's a HARD fact that bandwidth isn't growing as fast as storage and compute. That fact alone suggests heavy decentralization of service architectures in the coming years.

eitally 11y ago

All of the points subsequent to your first one conflict with that conjecture, that we'll see a huge swing back toward private cloud & on-premise DCs. Speaking from where I saw the industry [when I was a part of it] and now from what kinds of CIO/CTO conversations I see the GCP AMs/SEs having, I don't think we're remotely close to a swing back. Enterprise Capex budgets move VERY slowly most of the time, and it will take a multiple years to alter current momentum, which has finally reached the point where it's almost a foregone conclusion for many industries that on-prem DCs no longer make sense. This opinion is driven almost as much by the increasingly compelling portfolio of enterprisy SaaS products like Netsuite and a plethora of workforce management, CRMs, analytics/visualization, sooooo much more of what has historically been internally developed business productivity apps.

kordless 11y ago
There are a lot of people who think public cloud is the bee's knees and is going to continue to grow without bounds. The fact is, private cloud revenue is 30X as big as public cloud. It's this big because of previous growth cycles. Also, there's the consumer market to consider, just like in previous cycles. Public cloud is simply multi tenant computing on data managed by someone else away from where the customer created it. I'm predicting data will come home to roost on single tenant hardware (which comes in by way of IoT) and the code that has been called SaaS will follow it there. We'll still have managed services, they'll just run near us, instead of on compute owned by someone else. Regardless of whether you agree with that or not, it's a HARD fact that bandwidth isn't growing as fast as storage and compute. That fact alone suggests heavy decentralization of service architectures in the coming years.