429 karma · joined August 16, 2012
There was a separate testing client that was used for a lot of the demos. The demos on the 'real system' would fail so often that somebody came up with the idea to just take a lot of the code from the microservices and drop it straight alongside the testing client, as a monolith.
It was so much more stable that after becoming the default way to demo, it became the default way to use the tool.
5 years on and I believe it's still running that way. In this case microservices were massive over engineering.
They wrote two white papers which gave me a much better understanding of the challenges involved in creating a "market for compute" although I'll admit that much of the maths went over my head.
Mining economics: https://akash-web-prod.s3.amazonaws.com/uploads/2020/03/akas...
Decentralised cloud infrastructure marketplace: https://akash-web-prod.s3.amazonaws.com/uploads/2020/03/akas...
Awesome bare metal is a new repo created by Alex Ellis that tracks a lot of the projects: https://github.com/alexellis/awesome-baremetal
Also we (Packet) just open sourced Tinkerbell, our bare metal provisioning engine: https://www.packet.com/blog/open-sourcing-tinkerbell/
1. All collaboration requires that collaborators are able to recreate a shared version of reality
2. This means version controlling all the things
3. For 'normal' software teams it's often ok to do this for just code and environment, hence git + docker
4. But for data science teams they need to worry about more variables; code, environment, training + test data, hyper-parameters, summery statistics...
GitLFS allows teams to track training and test data (up to 2GB unless you run your own server IIRC) which removes a lot of the headaches around building tooling to tie all these variables together with, for example, Git + Docker + S3.
Dotscience.com is a good example of a project trying to solve this neatly.
Disclaimer: I used to work there.
Regarding the efficacy of online advertising, some businesses do notice but at the same time there's a larger question which is the trade-off between what online ads can achieve for different products at different stages.
Imagine a scale from "sitting in front of someone and explaining the value prop" to "an adwords ad."
If you're trying to remind someone about a brand they already know about, or attract them to a value prop that is easily explained online ads are great. If you have an unknown or hard to explain value prop you'd be better off starting elsewhere.
I agree that we need more research. Also, some suggested reading for anyone who's interested:
Traction: https://www.goodreads.com/book/show/22091581-traction (Good comparison of different marketing channels)
Scientific Advertising: https://www.goodreads.com/book/show/2621927-scientific-adver... (1923 book on measuring marketing)
Positioning: https://www.goodreads.com/book/show/760025.Positioning (First book to properly capture the ideas behind "brand marketing")
I did a talk[1] a few years back about my experiences helping a team to improve their communication skills. What surprised me was the emotional complications that prevent people from communicating effectively.
My guess is that the only way to get that is practice. Bring on the bad clients.
I don’t know anything else about them other than I think they are based in Europe. Might be worth a look.
Knowing which context you're in provides a useful tool when choosing mental models to approach the problem.
Having said that, 99% of mental models is practice as adopting a mental model is often an emotional problem too.
For example, while Cynefin may suggest an iterative approach for complex domains, you'll still need to be emotionally able to deal with the natural outcomes of iterative problem solving
e.g. being comfortable making decisions with incomplete data, handling failure, communicating progress when there is no clear way to estimate scope.
Furthermore, one way to avoid ending up in a feature for feature fight to the death with a bigger player is to think instead in terms of “job to be done”: https://hbr.org/2016/09/know-your-customers-jobs-to-be-done
By planning what you work on by focusing on what the customer is trying to achieve you’ll do yourself a favour when it comes to marketing as you’ll be able to speak in terms they understand.
Lastly, as you’ll see in the positioning book above, it’s not wise to completely avoid competitor analysis despite how unpleasant and draining it can sometimes be. A potential customer needs to understand why they would choose you over the incumbent. You need to know that answer before they can.