Discover and read the best of Twitter Threads about #technicaldebt
Most recents (6)
1/ We have been debating with some colleagues about the importance of learning about our decisions within the organization, and how the fast rotation of talent that's happening lately is affecting the decision-making outcomes quality. 
2/ Here we had the supposition that the feedback cycle between a decision is made and understanding the consequences are long enough. We always thrive for fast feedback loops, yet we acknowledge that that's not always possible.
3/ It's not the same doing TDD within a unit test - feedback cycle of seconds-minutes vs a business decision and go-to-market strategy that can take months.
We are talking about the latter. Where those business learnings are most valuable for the organization
We are talking about the latter. Where those business learnings are most valuable for the organization
I woke up to a great twitter thread from @QuinnyPig. As always, @QuinnyPig has a good twitter thread on mechanics and some technical details on what it means when @awscloud says it is will boot Parler off of #AWS
2 points: WHAT and HOW you build on #Cloud
2 points: WHAT and HOW you build on #Cloud
Two points are most important for any business that builds on Cloud: (1) WHAT is built on Cloud and (2) HOW it is built on it?
There are some great points here and wanted to start a twitter thread on my own quoting @QuinnyPig.
There are some great points here and wanted to start a twitter thread on my own quoting @QuinnyPig.
Quoting @QuinnyPig (emphasis added)...
"Making WHAT you built on AWS's systems work elsewhere is super challenging."
This making it "elsewhere" whether another cloud or on-premise servers requires how the original technical architecture and solution(s) are built.
"Making WHAT you built on AWS's systems work elsewhere is super challenging."
This making it "elsewhere" whether another cloud or on-premise servers requires how the original technical architecture and solution(s) are built.
for that first iteration of the code, simple don't mean shortcut:
simple != easy
simple != hacky
simple != rushed
simple != poorly written
even the first cut, MVP, needs to be changeable, testable, clear, expressive, etc
#TDD #technicalDebt #refactoring
simple != easy
simple != hacky
simple != rushed
simple != poorly written
even the first cut, MVP, needs to be changeable, testable, clear, expressive, etc
#TDD #technicalDebt #refactoring
the "make it work" part (K.Beck) always gets mistaken to mean, hack at the code, squeeze, force-fit, finagle, the easiest and quickest thing that'll get the managers, customers, users off our back.
but we just end up on the wrong side of karma
2/3
but we just end up on the wrong side of karma
2/3
it just adds high interest technical debt, and need to refactor out:
- initial understanding, with deeper insight into the feature
- tightly coupled, low cohesion, incoherent, viscous, rigid, fragile, immobile code
#highInterestTechnicalDebt
#technicalDebt
3/3
- initial understanding, with deeper insight into the feature
- tightly coupled, low cohesion, incoherent, viscous, rigid, fragile, immobile code
#highInterestTechnicalDebt
#technicalDebt
3/3
#CodeSmell 01: Your objects are a bunch of public attributes
Problem: en.wikipedia.org/wiki/Anemic_do…
Solutions:
1) Find Responsibilities.
2) Protect your attributes.
3) Hide implementations
4) Delegate
Problem: en.wikipedia.org/wiki/Anemic_do…
Solutions:
1) Find Responsibilities.
2) Protect your attributes.
3) Hide implementations
4) Delegate
#CodeSmell 02: Code has lots of comments
Problem: Comments are coupled to implementation
Solutions:
1) Refactor methods
2) Rename methods to more declarative unes
3) break methods
4) If a comment describe what a method does. Just name the method with this description
Problem: Comments are coupled to implementation
Solutions:
1) Refactor methods
2) Rename methods to more declarative unes
3) break methods
4) If a comment describe what a method does. Just name the method with this description
Fionta and ECP talk about caring for a large website. #19ntcsustainweb #19ntc
@EnterpriseNow is a large #Drupal website but session takeaways are for all.
Know your site’s goals and track them. Know your audience and their goals. How do you quantify success? #19ntcsustainweb
Getting started with Want to succeed with #TDD ? Stop writing tests! @BurkHufnagel #TDD #DevNexus devnexus.com/presentations/…
We work with software that's way more complex than a simple shape like this. Reasonable people may see the same thing in different ways, based on their perspective! @BurkHufnagel #TDD #DevNexus
You have a goal that you share with me: you want delivering better code faster
... not *writing* code more quickly! @BurkHufnagel #TDD #DevNexus
... not *writing* code more quickly! @BurkHufnagel #TDD #DevNexus
