Hacker News new | past | comments | ask | show | jobs | submit login

For about a decade of my life I was looked down upon by engineers because I made advertising that targeted engineers :-) I agree with a lot of what is in this article, there's one thing missing:

Highlight differences. Especially those that are quality of life improvement for the engineering organization.

AWS was weird (literally - it was very unique) and slower than dedicated servers when it came out. But there was a difference, and AWS let me know: No need to talk to people to provision, and once you activate an instance it will be available in minutes. That alone was enough to switch from on prem or managed colo resources.




The first database server I bought required 3 months of negotiations and approvals, multiple vendor meetings, ~100 sheets of paper, 4 hours of capital asset tracking paperwork, and annual meetings with auditors to show them where the expensive metal box was.

My last database server just had to comply with a tagging policy.

That's what prompted me to switch (for work, anyway).


This is one of the reasons why I disagree with point three of the article.

The emotion that AWS helped overcome was frustration that individual developers faced when trying to build something new. Suddenly, hardware was in their control from their keyboards.

That was a magical experience, and it definitely filled me with emotion the first time I uploaded an object to S3.

(I loved it so much, I later worked for that team!)


people extol the benefits of cloud for fast scaling, powerful managed services etc and those things are absolutely true. But in my mind, the real "value add" in many industries was not the services. It was the API...


I just remember being in a meeting about colo and on prom where people argued about why the colo was better... I provisioned a working instance of our web site (a big job board) with load balancer and fresh databases via the API... and the arguments stopped when I showed the team. "While you were arguing, I deployed our entire infrastructure into AWS. Try it out at http://somewierdawsurl.something"


These are nice. Still IMO the real value was moving from the CAPEX column to the OPEX column.


Because outsourced CAPEX become OPEX. Virtual bro-fist!

Funny thing though: Apple still has high amounts of CAPEX. I guess they buy their machinery at Foxconn.


Difficult to look down upon people who are positioning themselves above others (know what you need better than yourself, power of influencing into what they sell, revenue/benefit ratio, deceiving undetected, etc.), but this way when we look up all that we see are a bunch of asses. ; )


Another thing AWS did that was really out there back then was to allow on-the-go, only pay for what you use workloads. EC2 instances were billed per hour instead of per month.


There might be a difference between engineers, admins and all the people who want to get work done and are tired of technical gatekeepers.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: