Saul - first thank you so much for your post. I am learning a lot. I mistakenly have known about this board for about 6 years and have not pulled the trigger on reading the knowledgebase til now. I was too focussed on trading which takes up too much time and I am not successful at it. Had I followed the recommendations here, I would be much richer. Live and learn. I hope you are around for many more years for us all to learn more.
Saul/brittlerock
But it is these relative decisions that set us apart. Let me explain. During my career I watched two database products dominate the industry. First, the IBM product, IMS DB/TP (that translates to Information Management System Database/Teleprocessing). IMS was not the only database product
…
Why am I relating all this history? In a word, Mongo.
I want place an addendum here and say I believe you are both correct in the potential success of Mongo but for other reasons.
I work in Software Engineering and have done so for 20+ years. I started coding large enterprise systems in Java and over the years have learned lots of different languages (JavaScript, Scala, Kotlin etc.) and now lead teams of engineerings on high throughput, low latency, critical systems.
My experience of Mongo as a company (10Gen a few years ago) has been appalling - they tried to modify our NDA during evaluation stage to take any IP of new products they discovered while working with us and removed any of our rights to those products. It leaves a bitter taste still.
At the same time, there was an explosion by an anonymous blog post stating everything wrong with MongoDB including cascading failures, faulty data integrity (transactions reporting as completed when they weren’t) etc. All 10Gen’s CTO could say was “what are the support ticket numbers?” A pathetic response, at best. For those of you that don’t know, data integrity is key in any system. Without it, we just get the wrong answers and tracking down WHY is extremely difficult.
Now let’s turn to today. Mongo are going gangbusters. They now have transactions spanning clusters and I have found on their own JIRA board with just a few seconds of searching a couple of bugs submitted saying it is terrible at best.
BUT
this does not matter.
IMO, IT execs look at features such as cloud agnostic, Atlas, full text search and any other products that will get them up and running quickly for a fast route to market/solution to their issues. They want to alleviate their pain quickly.
If the Mongo issues I described emerge, by that point there is a general consensus in the company of “the engineers have to fix it” and the Mongo contract has already been signed.
They clearly have a fantastic sales arm and product strategy. From an engineer’s POV, their tech execution leaves us wanting.
The sales decision is disjoined from the production performance of the system the Mongo products are used in but it doesn’t matter.
The metrics say it all and I plan on opening a position in Mongo too.
Thanks again for everyone’s contributions.
Much appreciated.