“As we move into AI first development, we must design systems which encourage data transfer between systems.” (via @Pocket) #longreads https://medium.com/@damonallison/evolving-microservices-with-event-sourcing-7396e015cfd2
Reading “Pipeline Jungles” from https://papers.nips.cc/paper/5656-hidden-technical-debt-in-machine-learning-systems.pdf
Welcome to the pipeline jungle, we got data and queries
We got everything you want, we know the schemes
We are the scientists that can find whatever you may need
If you got the compute, we got your forecasted time series
“It is important to create team cultures that reward deletion of features, reduction of complexity, improvements in reproducibility, stability, and monitoring to the same degree that improvements in accuracy are valued.”
“The hardest part of machine learning today is actually deploying and maintaining accurate models, as it requires constant access to new data to update them and improve their accuracy.”
“State Is the Hardest Part and the Most Interesting Opportunity for Serverless” (via @Pocket) #longreads https://thenewstack.io/serverless-needs-a-bolder-stateful-vision/
“One way sponsors can break the mini-me pattern is to prioritize difference — perhaps a unique skillset or distinct personal brand — as they consider potential protégés.” (via @Pocket) https://hbr.org/2019/02/sponsors-need-to-stop-acting-like-mentors
The benefits of the practice and feedback loops of CI transcend the specifics. They make it easier to be more CloudFirst over time (if you need/want to).
Applies equally to VMs and firmware (and anything else): https://itrevolution.com/the-amazing-devops-transformation-of-the-hp-laserjet-firmware-team-gary-gruver/ https://twitter.com/fintanr/status/1103244723538616320
Nodding my head as Screaming in the Cloud talks about the ineffectiveness of “yelling at humans.” Another quality @QuinnyPig podcast. 👍 https://overcast.fm/+MdF6j0abg
“The same is true for machine learning: from the beginning, it’s important to incorporate security experts and domain experts who understand how a system is likely to be abused.”
You created a machine learning application. Now make sure it’s secure.
“Data science is hard to productionize, and one of the reasons it is hard is because it has so many moving parts.” (via @Pocket) #longreads https://emilygorcenski.com/post/data-versioning/
🎶After 12 outages, I need Gremlin
Feed me chaos and I start trembling
The thrill of suspense is intense, you’re horrified
But this ain’t the cinemas or Tales from the Darkside 🎶
Yep. There are definitely *migration costs* associated with changing things up. Whether those are so high (financial, opportunity, cognitive, etc.) to prevent an immediate change is a nuanced conversation.
We’re already locked into the latent assumptions that go unquestioned.
“Meanwhile, DEV will have to coordinate every little move they take with OPS.” (via @Pocket) https://medium.com/@whoisziv/dear-deployment-diary-serverless-is-fucking-hard-4d5f409347af
“Serverless has a bright future, but only when it provides the same productivity as it does cost value.” (via @Pocket) https://medium.freecodecamp.org/serverless-doesnt-have-to-be-an-infuriating-black-box-b23cca2b2ba2
“At the end of the day, if you’re an asshole, you’re a failure as a human being because you promote unnecessary suffering. What else is there to say?” (via @Pocket) #longreads https://www.vox.com/conversations/2017/9/26/16345476/stanford-psychologist-art-of-avoiding-assholes
“Executing ML inference using a full serverless approach can be very effective. If the execution time can be limited and the right batch size is used, it is more affordable than a permanently hosted model and endpoint.” (via @Pocket) https://medium.com/merapar/pure-serverless-machine-learning-inference-with-aws-lambda-and-layers-979702d9ae49
“We must say no to things because there are always too many features to build, too many directions we can take, and too many policies we can adopt.” (via @Pocket) https://thoughtbot.com/blog/say-no-to-more-process-say-yes-to-trust
“The hardest thing to do is to trust in a process though.” (via @Pocket) https://medium.com/@PaulDJohnston/we-need-to-stop-calling-them-sprints-couch-to-5k-is-the-right-analogy-22eb41729acf
“Unfortunately, since June 2018, we have witnessed significant intermingling of proprietary code into the code base.” (via @Pocket) #longreads https://aws.amazon.com/blogs/opensource/keeping-open-source-open-open-distro-for-elasticsearch/
“Outside of some managed offerings, Apache Pulsar (The distributed pub-sub and queuing system) with Pulsar Functions was the simplest topology.” (via @Pocket) https://www.jowanza.com/blog/2019/3/9/efficient-stream-processing-with-pulsar-functions
X: We can fix this with a new policy
Me: The Amazon delivery person literally left the packages directly beneath a sign that says “Do not leave packages here.”
“Buy rather than build, unless it’s critical to your business.” is just one of the many excellent quotes in @sarahjwells preso: https://speakerdeck.com/sarahjwells/qcon-london-2019-mature-microservices-and-how-to-operate-them
I’m all for writing more secure services, but the notion that #serverless introduces fundamentally new and different risks is something I don’t quite understand.
“Given that coordination and communication swamp all other costs in modern software development it is a pressing area to invest in...” (via @Pocket) https://kellanem.com/notes/new-tech
“It feels like we’re starting to pass the peak of the hype cycle of microservices.” (via @Pocket) http://www.craigkerstiens.com/2019/03/13/give-me-back-my-monolith/
“Which is more likely, that people working at these companies don’t know what they’re doing or people don’t know what these companies are doing?”
Stuff The Internet Says On Scalability For March 8th, 2019 (via @Pocket) #longreads http://highscalability.com/blog/2019/3/8/stuff-the-internet-says-on-scalability-for-march-8th-2019.html
“People debate the information you put in front of them. Never put information in front of people if you don’t want it discussed.”
How to Prepare for a Board Meeting to Make Sure you Crush It (via @Pocket) #longreads https://bothsidesofthetable.com/how-to-prepare-for-a-board-meeting-to-make-sure-you-crush-it-b52b8ce61636
“The state of Kubernetes config management has never been more exciting” (via @Pocket) #longreads https://blog.argoproj.io/the-state-of-kubernetes-configuration-management-d8b06c1205
“Do a great job and you will be ignored.” (via @Pocket) https://medium.com/adobetech/why-i-a-serverless-developer-dont-care-about-your-containers-40c08d36aee4
“We have chosen to organize Computer Science research differently at Google by maximally connecting research and development.” (via @Pocket) #longreads http://norvig.com/cacm-hybrid.html
“Are many of the modern frontend tools and practices just technical dept in disguise?”
The “Backendification” of Frontend Development (via @Pocket) #longreads https://medium.com/@mktlr/the-backendification-of-frontend-development-62f218a773d4
True. While I could have been vulnerable to that same symptom with elastic cloud resources today, it’s less likely just because of the work involved to set everything up IME.
#serverless makes it much more prominent from Day one.
I don’t like or give whiteboard interviews. My preferred interview is basically “Explain to me something you know well while I occasionally ask questions.” https://twitter.com/jboursiquot/status/1108113474008346636
RT @perrito666: Interviews with matt are super cool :) i had one and it was a very enjoyable experience https://twitter.com/mweagle/status/1108176453131223040
“If your team decides to focus only on the value it’s delivering, and delegate anything outside that either to another team, or ideally outside — then your team is going serverless. “ (via @Pocket) #longreads https://read.acloud.guru/serverless-is-a-state-of-mind-717ef2088b42
“I made a little flow chart of mainstream programming languages and how programmers seem to move from one to another.” https://apenwarr.ca/log/20190318
🧐 “The growing adoption of serverless is fuelled by the increased proliferation of container-based applications which are cloud-native — an architecture that’s required for Serverless.” (via @Pocket) https://www.itproportal.com/features/five-trends-in-serverless-for-2019/
“The Amazon and Elastic controversy is the product of a collision of models.” (via @Pocket) #longreads https://redmonk.com/sogrady/2019/03/15/cloud-open-source-powder-keg/
“Ouch. Variables. Qualified names. Arguments. This is not structured data. This is programming masquerading as configuration.” (via @Pocket) https://blog.atomist.com/in-defense-of-yaml/
“To start addressing the limitations of relational databases, we reconceptualized the stack by decomposing the system into its fundamental building blocks.”
Amazon Aurora ascendant : How we designed a cloud-native relational database (via @Pocket) https://www.allthingsdistributed.com/2019/03/Amazon-Aurora-design-cloud-native-relational-database.html?utm_campaign=newsletter_subscription&utm_medium=email&utm_source=nuzzel
“Any sustainable machine learning practice must address machine learning’s unique security issues.”
You created a machine learning application. Now make sure it’s secure. (via @Pocket) #longreads https://www.oreilly.com/ideas/you-created-a-machine-learning-application-now-make-sure-its-secure
“The reasoning is that for any function you can think of, at least some applications will find that by necessity they must implement the function themselves in order to meet correctly their own requirements.”
“Maybe some truisms no longer apply at some system change complexity inflection point. Truism 1: Complex systems evolve from simpler systems. Truism 2: Rebuilding a new system from scratch is riskier than changing an existing system.” http://highscalability.com/blog/2019/3/22/stuff-the-internet-says-on-scalability-for-march-22nd-2019.html
“As you’re laying the foundation for an event-driven architecture, following a few rules can help keep things manageable.” (via @Pocket) https://rjzaworski.com/2019/03/7-commandments-for-event-driven-architecture
“If you aren’t sure what to instrument, imagine the system being completely overloaded and slow: what parameters would we need to be able to find and fix the problem?” (via @Pocket) #longreads https://queue.acm.org/detail.cfm?id=1921361
Suggested response when you review a co-worker’s PR that preserves functionality *and* reduces LOC. https://twitter.com/Kazeem/status/1109270192629170176
@aneel: “In every SaaS company’s life there’s a moment where you have to choose to not take a customer because that customer will probably destroy your company.” https://overcast.fm/+RrQWKeC0
“I’ve come to believe that every system implicitly defines a spectrum of changes, ordered by their likelihood.” (via @Pocket) http://www.michaelnygard.com/blog/2017/06/spectrum-of-change
“This was not supposed to be a story about burnout, this was supposed to be the things I learned working through it and being able to see the other side.” (via @Pocket) #longreads https://cate.blog/2019/02/06/the-cost-of-fixing-things/
“If you think women in tech is just a pipeline problem, you haven’t been paying attention.” (via @Pocket) #longreads https://medium.com/tech-diversity-files/if-you-think-women-in-tech-is-just-a-pipeline-problem-you-haven-t-been-paying-attention-cb7a2073b996
“I firmly ascribe to the dictum that you don’t let your tools dictate your process. I am also a firm believer in using a physical board for a team to organize work.”
Agile at AstrumU: A Case Study (via @Pocket) #longreads https://blog.kevingoldsmith.com/2019/02/04/agile-at-astrumu-a-case-study/
“Gmail doesn’t sort your emails by the date the way Inbox did. Happily, you can re-create the feature yourself with a little bit of hackery.” (via @Pocket) https://www.howtogeek.com/366260/group-your-gmail-inbox-by-date-like-google-inbox-did-and-outlook-does/
“Kinesis shards can collect up to 1 megabyte per second of data at 1,000+ transactions per second. Amazon Kinesis apps can read data from each shard at up to 2 megabytes per second.” (via @Pocket) https://www.jenunderwood.com/2017/04/03/amazon-kinesis-for-streaming-data-analytics/
“When most people think about the perils of the sea, they think about sharks, electric eels or perhaps Ursula from The Little Mermaid – my concerns are a little different.” (via @Pocket) #longreads https://exchange.telstra.com.au/the-complexities-and-quirks-of-protecting-our-subsea-cables/
“The truth, however, is that a busy schedule alone isn’t enough to trigger professional burnout.” (via @Pocket) https://www.forbes.com/sites/johnrampton/2015/05/13/the-6-causes-of-professional-burnout-and-how-to-avoid-them/#746898a01dde
“With data science, you learn as you go, not before you go.”
(via @Pocket) #longreads https://hbr.org/2019/03/why-data-science-teams-need-generalists-not-specialists
“At this point you’ve finished your organizational design, and there is only one commandment left to master: stop making organizational changes.” (via @Pocket) #longreads https://lethain.com/how-to-evolve-eng-org/
