SlideShare a Scribd company logo
1 of 43
1
​Renovating and maintaining
digital services and data
​Lessons from traditional infrastructure
​Ade Adewunmi
​@adewunmi
​October 24th
2018
2
Hi
3
Great power, great responsibility
4
I’ve been thinking a lot about technological
utopianism, agile delivery processes and the
focus on the ‘happy path’ and how these things
shape the way we think, work and the things we
build.
5
H0
: Technologists are no more optimistic than
the wider population.
H1
: Technologists are more optimistic than the
rest of the population.
I’ve also been thinking about whether there’s a
type of person who’s drawn to work in data and
digital.
6
I haven’t quite worked out a suitably rigorous way
of testing the null hypothesis (so I don’t have the
evidence to reject it).
7
I haven’t quite worked out a suitably rigorous way
of testing the null hypothesis (so I don’t have the
evidence to reject it).
But the alternative hypothesis is one of my working
assumptions (based on my personal experience).
8
Some other working hypotheses/assumptions:
9
Some other working hypotheses and assumptions:
A technocratic focus often creates blind spots.
Technologists are often blind to the economic,
social and power dynamics at play, in human
interactions.
10
OK, so there’s a lot to trip us up. So what?
11
OK, so there’s a lot to trip us up. So what?
So, exploring structural obstacles to good
renovation and maintenance culture by exploring
failures in infrastructure building is more useful
than studying successes.
12
A purely technocratic approach to building data
infrastructure that ignores the drivers shaping
the environments in which our infrastructure is
deployed and built, results in brittle
infrastructure.
13
This isn’t the
cheeriest start but
stick with me, I’m
going for balanced
and hopeful.
(honestly)
14
We share many traits with other technocrats
including builders and designers of traditional
infrastructure.
We’re likely to make similar mistakes which means
we can also learn from these builders and
designers.
15
Lessons from traditional
infrastructure
©2018 Teradata
16
Factors that shape our attitude to renovating
and maintaining infrastructure:
• economic
17
Factors that shape our attitude to renovating and
maintaining infrastructure:
• economic
• social
18
Factors that shape our attitude to renovating and
maintaining infrastructure
• economic
• social
• power dynamics
19
In the slides that follow, I reference the NYC metro
system (MTA) as a way of illustrating some points I want
to make.
Disclaimer: I’m not an expert on the NYC MTA and the
CityLab article was written in April of this year so things
might have changed since then.
CityLab article:
20
December 16, 1940: date
the last new subway line
was opened, aside from a
handful of small extensions
and connections.
Source: CityLab article.
21
December 16, 1940.
3 contributing factors:
1. Lure of the suburbs
22
December 16, 1940.
3 contributing factors:
1. Lure of the suburbs
2. Delivery partner
challenges
23
December 16, 1940.
3 contributing factors:
1. Lure of the suburbs
2. Delivery partner
challenges
3. Growing cost of
technical debt
24
The lure of the suburbs, provides a brilliant lens for
viewing (and understanding) the defunding of the
‘infrastructure commons’.
Geographic distance means people have less in
common - and that has economic and social
implications.
25
Funding choices shapes the type of infrastructure we get.
For example, the City of Chicago’s decision to sell off the
management and revenue streams from its parking
metres led to a big spike in metre fees and a decline in
quality of service.
Closer to home, HMT’s funding process affects the shape
of the UK government’s digital infrastructure too.
26
Code reuse in government is still not widely
practiced and data reuse is even further behind.
“Reusable code has to be specifically designed
for a generalised purpose and it is unlikely to
appear spontaneously as a natural by-product of
development projects.”Ben Morris blog.
27
Culture shapes infrastructure; infrastructure shapes
culture: does your infrastructure support reuse?
Data scientists working in
government departments,
making models
Professionals within a data
federation/ecosystem using
trained model
Exploration
•Data Wrangling
•DS Lab
•Model scripting (untrained
models)
•Testing, Training, Model
Evaluation
•Version Control
•Dependency Management
Automate
• Software unit tests
• Model Training
• Storage of trained models
• Model Evaluation
• Model Business
Approval/Report Creation
• Comparison vs current
Live model
(Champion/Challenger)
Consume
•Real-time model scoring
engines
•Automatic deployment of
trained model artefacts
•Dashboards and forecasts
updated using new models
•Model performance
monitoring
•Model output logging
Involving: Analysts, Data Scientists, Engineers, Dev Ops, Business Stakeholders
© 2018 Teradata
28
Culture shapes infrastructure; infrastructure shapes
culture
Data Scientist
develops model
❑ Models developed by data scientist using preferred, locally installed analytic
tools
Data Scientist
prepares
power point
❑ A data scientist prepares a Powerpoint presentation on the benefits of a new
model
❑ Manually creating standard model performance metrics of model
Model approval
committee
❑ Model business approval committee meeting to:
See presentation; Discuss benefits/risks; Approve/Reject model
Model manually
transferred to
relevant production
environment
❑ Once approved, model exported to an Excel file of weights
❑ Excel manually transcribed into format suitable for the Production environment
running on organisation-specific systems
Model
goes live
❑ New model goes live in production under close supervision
© 2018 Teradata
29
Culture shapes infrastructure; infrastructure shapes
culture
Data Scientist
develops model
❑ Models developed in standardised Python or R
Automated
tests
❑ Model is tested using a validation data set
❑ Report with model performance metrics is automatically generated
Peer
approval
❑ Model performance report is available in a UI for approval
❑ Approving in the UI will trigger automatically the bundling of models into
business use case
Business
approval
❑ The business impact of the new bundle of models is estimated and displayed
in a UI for the business team to approve the bundle
Model
goes live
❑ New model gets deployed automatically with A/B testing capability to validate
business impact
❑ The performance of the models in production are continuously monitored
© 2018 Teradata
30
Traditional productivity
tools assume highly
individualised, localised
ways of working.
Culture shapes infrastructure; infrastructure shapes
culture: do your tools support re-use?
31
How we balance investment in paying down
technical debt against building new infrastructure
is not a neutral decision.
It directly contributes to whether or not we develop a
healthy maintenance culture.
How much extra work are we willing to do, to reuse
and create reusable data?
32
Room for improvement
33
• Build stronger links between design and policy, using
data
Culture shapes infrastructure, infrastructure
shapes culture
34
Policy is, still, too often
made in an evidence
free vacuum.
Good data is one of
the ways we help
shine a light on policy
gaps.
35
• Build stronger links between design and policy, using
data
• Apply serious effort to incorporating responsible design
of the data lifecycle - data discovery, data collection,
metadata generation - into design sprints
Culture shapes infrastructure, infrastructure
shapes culture
36
We have to be aware of the
challenges and risks of
indiscriminate data gathering.
And acknowledge the power
dynamics at play and inherent
biases in data collection.
Thinking beyond the ‘happy
path’
37
38
• Build stronger links between design and policy, using
data
• Apply serious effort to incorporating responsible design
of the data lifecycle - data discovery, data collection,
metadata generation - into design sprints
• Standardise and reward data reuse (both supply and
demand)
Culture shapes infrastructure, infrastructure
shapes culture (part I)
39
• Think more critically about whether our tools and
working environment (e.g. Google Analytics)
broaden our view of the data ecosystem or narrows it
• Raise the threshold of ‘good enough’ data quality
(defining it would be a good start!)
• Build a strong data commons by building a wide
community of users
Culture shapes infrastructure, infrastructure
shapes culture (part II)
40
• Let’s think about the stories we tell - what are we
emphasising and celebrating?
Culture shapes infrastructure, infrastructure
shapes culture (part III)
41
Reasons for hope
©2018 Teradata
42
• We’re waking up to the brittleness of our infrastructure.
• We’ve been here before. There’s infrastructure (social
and software) from the digital transformation era we can
leverage.
• Efforts to build community and maintain a digital
commons persist.
Why I remain hopeful about the future of
government infrastructure
43
Thank you.
©2018 Teradata
Thank you.
©2018 Teradata

More Related Content

Recently uploaded

Call Girls In datia Escorts ☎️7427069034 🔝 💃 Enjoy 24/7 Escort Service Enjoy...
Call Girls In datia Escorts ☎️7427069034  🔝 💃 Enjoy 24/7 Escort Service Enjoy...Call Girls In datia Escorts ☎️7427069034  🔝 💃 Enjoy 24/7 Escort Service Enjoy...
Call Girls In datia Escorts ☎️7427069034 🔝 💃 Enjoy 24/7 Escort Service Enjoy...
nehasharma67844
 
Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...
Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...
Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...
Chandigarh Call girls 9053900678 Call girls in Chandigarh
 
call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...
call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...
call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 

Recently uploaded (20)

VIP Model Call Girls Shikrapur ( Pune ) Call ON 8005736733 Starting From 5K t...
VIP Model Call Girls Shikrapur ( Pune ) Call ON 8005736733 Starting From 5K t...VIP Model Call Girls Shikrapur ( Pune ) Call ON 8005736733 Starting From 5K t...
VIP Model Call Girls Shikrapur ( Pune ) Call ON 8005736733 Starting From 5K t...
 
Call Girls In datia Escorts ☎️7427069034 🔝 💃 Enjoy 24/7 Escort Service Enjoy...
Call Girls In datia Escorts ☎️7427069034  🔝 💃 Enjoy 24/7 Escort Service Enjoy...Call Girls In datia Escorts ☎️7427069034  🔝 💃 Enjoy 24/7 Escort Service Enjoy...
Call Girls In datia Escorts ☎️7427069034 🔝 💃 Enjoy 24/7 Escort Service Enjoy...
 
Call On 6297143586 Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
Call On 6297143586  Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...Call On 6297143586  Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
Call On 6297143586 Yerwada Call Girls In All Pune 24/7 Provide Call With Bes...
 
Junnar ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For S...
Junnar ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For S...Junnar ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For S...
Junnar ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For S...
 
best call girls in Pune - 450+ Call Girl Cash Payment 8005736733 Neha Thakur
best call girls in Pune - 450+ Call Girl Cash Payment 8005736733 Neha Thakurbest call girls in Pune - 450+ Call Girl Cash Payment 8005736733 Neha Thakur
best call girls in Pune - 450+ Call Girl Cash Payment 8005736733 Neha Thakur
 
Call On 6297143586 Viman Nagar Call Girls In All Pune 24/7 Provide Call With...
Call On 6297143586  Viman Nagar Call Girls In All Pune 24/7 Provide Call With...Call On 6297143586  Viman Nagar Call Girls In All Pune 24/7 Provide Call With...
Call On 6297143586 Viman Nagar Call Girls In All Pune 24/7 Provide Call With...
 
VIP Model Call Girls Narhe ( Pune ) Call ON 8005736733 Starting From 5K to 25...
VIP Model Call Girls Narhe ( Pune ) Call ON 8005736733 Starting From 5K to 25...VIP Model Call Girls Narhe ( Pune ) Call ON 8005736733 Starting From 5K to 25...
VIP Model Call Girls Narhe ( Pune ) Call ON 8005736733 Starting From 5K to 25...
 
Pimpri Chinchwad ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi R...
Pimpri Chinchwad ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi R...Pimpri Chinchwad ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi R...
Pimpri Chinchwad ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi R...
 
Election 2024 Presiding Duty Keypoints_01.pdf
Election 2024 Presiding Duty Keypoints_01.pdfElection 2024 Presiding Duty Keypoints_01.pdf
Election 2024 Presiding Duty Keypoints_01.pdf
 
Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...
Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...
Russian🍌Dazzling Hottie Get☎️ 9053900678 ☎️call girl In Chandigarh By Chandig...
 
Pimple Gurav ) Call Girls Service Pune | 8005736733 Independent Escorts & Dat...
Pimple Gurav ) Call Girls Service Pune | 8005736733 Independent Escorts & Dat...Pimple Gurav ) Call Girls Service Pune | 8005736733 Independent Escorts & Dat...
Pimple Gurav ) Call Girls Service Pune | 8005736733 Independent Escorts & Dat...
 
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Chakan Call Me 7737669865 Budget Friendly No Advance Booking
 
Hinjewadi * VIP Call Girls Pune | Whatsapp No 8005736733 VIP Escorts Service ...
Hinjewadi * VIP Call Girls Pune | Whatsapp No 8005736733 VIP Escorts Service ...Hinjewadi * VIP Call Girls Pune | Whatsapp No 8005736733 VIP Escorts Service ...
Hinjewadi * VIP Call Girls Pune | Whatsapp No 8005736733 VIP Escorts Service ...
 
1935 CONSTITUTION REPORT IN RIPH FINALLS
1935 CONSTITUTION REPORT IN RIPH FINALLS1935 CONSTITUTION REPORT IN RIPH FINALLS
1935 CONSTITUTION REPORT IN RIPH FINALLS
 
AHMR volume 10 number 1 January-April 2024
AHMR volume 10 number 1 January-April 2024AHMR volume 10 number 1 January-April 2024
AHMR volume 10 number 1 January-April 2024
 
Financing strategies for adaptation. Presentation for CANCC
Financing strategies for adaptation. Presentation for CANCCFinancing strategies for adaptation. Presentation for CANCC
Financing strategies for adaptation. Presentation for CANCC
 
Chakan ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For S...
Chakan ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For S...Chakan ( Call Girls ) Pune  6297143586  Hot Model With Sexy Bhabi Ready For S...
Chakan ( Call Girls ) Pune 6297143586 Hot Model With Sexy Bhabi Ready For S...
 
2024: The FAR, Federal Acquisition Regulations, Part 30
2024: The FAR, Federal Acquisition Regulations, Part 302024: The FAR, Federal Acquisition Regulations, Part 30
2024: The FAR, Federal Acquisition Regulations, Part 30
 
call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...
call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...
call girls in Raghubir Nagar (DELHI) 🔝 >༒9953056974 🔝 genuine Escort Service ...
 
celebrity 💋 Agra Escorts Just Dail 8250092165 service available anytime 24 hour
celebrity 💋 Agra Escorts Just Dail 8250092165 service available anytime 24 hourcelebrity 💋 Agra Escorts Just Dail 8250092165 service available anytime 24 hour
celebrity 💋 Agra Escorts Just Dail 8250092165 service available anytime 24 hour
 

Featured

Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie Insights
Kurio // The Social Media Age(ncy)
 
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them wellGood Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Saba Software
 
Introduction to C Programming Language
Introduction to C Programming LanguageIntroduction to C Programming Language
Introduction to C Programming Language
Simplilearn
 

Featured (20)

How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024
 
Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie Insights
 
Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024
 
5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary
 
ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd
 
Getting into the tech field. what next
Getting into the tech field. what next Getting into the tech field. what next
Getting into the tech field. what next
 
Google's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search IntentGoogle's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search Intent
 
How to have difficult conversations
How to have difficult conversations How to have difficult conversations
How to have difficult conversations
 
Introduction to Data Science
Introduction to Data ScienceIntroduction to Data Science
Introduction to Data Science
 
Time Management & Productivity - Best Practices
Time Management & Productivity -  Best PracticesTime Management & Productivity -  Best Practices
Time Management & Productivity - Best Practices
 
The six step guide to practical project management
The six step guide to practical project managementThe six step guide to practical project management
The six step guide to practical project management
 
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
 
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
 
12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work
 
ChatGPT webinar slides
ChatGPT webinar slidesChatGPT webinar slides
ChatGPT webinar slides
 
More than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike RoutesMore than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike Routes
 
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
 
Barbie - Brand Strategy Presentation
Barbie - Brand Strategy PresentationBarbie - Brand Strategy Presentation
Barbie - Brand Strategy Presentation
 
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them wellGood Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
 
Introduction to C Programming Language
Introduction to C Programming LanguageIntroduction to C Programming Language
Introduction to C Programming Language
 

Maintenance-by-design: Building resilient digital services for a data-driven world

  • 1. 1 ​Renovating and maintaining digital services and data ​Lessons from traditional infrastructure ​Ade Adewunmi ​@adewunmi ​October 24th 2018
  • 3. 3 Great power, great responsibility
  • 4. 4 I’ve been thinking a lot about technological utopianism, agile delivery processes and the focus on the ‘happy path’ and how these things shape the way we think, work and the things we build.
  • 5. 5 H0 : Technologists are no more optimistic than the wider population. H1 : Technologists are more optimistic than the rest of the population. I’ve also been thinking about whether there’s a type of person who’s drawn to work in data and digital.
  • 6. 6 I haven’t quite worked out a suitably rigorous way of testing the null hypothesis (so I don’t have the evidence to reject it).
  • 7. 7 I haven’t quite worked out a suitably rigorous way of testing the null hypothesis (so I don’t have the evidence to reject it). But the alternative hypothesis is one of my working assumptions (based on my personal experience).
  • 8. 8 Some other working hypotheses/assumptions:
  • 9. 9 Some other working hypotheses and assumptions: A technocratic focus often creates blind spots. Technologists are often blind to the economic, social and power dynamics at play, in human interactions.
  • 10. 10 OK, so there’s a lot to trip us up. So what?
  • 11. 11 OK, so there’s a lot to trip us up. So what? So, exploring structural obstacles to good renovation and maintenance culture by exploring failures in infrastructure building is more useful than studying successes.
  • 12. 12 A purely technocratic approach to building data infrastructure that ignores the drivers shaping the environments in which our infrastructure is deployed and built, results in brittle infrastructure.
  • 13. 13 This isn’t the cheeriest start but stick with me, I’m going for balanced and hopeful. (honestly)
  • 14. 14 We share many traits with other technocrats including builders and designers of traditional infrastructure. We’re likely to make similar mistakes which means we can also learn from these builders and designers.
  • 16. 16 Factors that shape our attitude to renovating and maintaining infrastructure: • economic
  • 17. 17 Factors that shape our attitude to renovating and maintaining infrastructure: • economic • social
  • 18. 18 Factors that shape our attitude to renovating and maintaining infrastructure • economic • social • power dynamics
  • 19. 19 In the slides that follow, I reference the NYC metro system (MTA) as a way of illustrating some points I want to make. Disclaimer: I’m not an expert on the NYC MTA and the CityLab article was written in April of this year so things might have changed since then. CityLab article:
  • 20. 20 December 16, 1940: date the last new subway line was opened, aside from a handful of small extensions and connections. Source: CityLab article.
  • 21. 21 December 16, 1940. 3 contributing factors: 1. Lure of the suburbs
  • 22. 22 December 16, 1940. 3 contributing factors: 1. Lure of the suburbs 2. Delivery partner challenges
  • 23. 23 December 16, 1940. 3 contributing factors: 1. Lure of the suburbs 2. Delivery partner challenges 3. Growing cost of technical debt
  • 24. 24 The lure of the suburbs, provides a brilliant lens for viewing (and understanding) the defunding of the ‘infrastructure commons’. Geographic distance means people have less in common - and that has economic and social implications.
  • 25. 25 Funding choices shapes the type of infrastructure we get. For example, the City of Chicago’s decision to sell off the management and revenue streams from its parking metres led to a big spike in metre fees and a decline in quality of service. Closer to home, HMT’s funding process affects the shape of the UK government’s digital infrastructure too.
  • 26. 26 Code reuse in government is still not widely practiced and data reuse is even further behind. “Reusable code has to be specifically designed for a generalised purpose and it is unlikely to appear spontaneously as a natural by-product of development projects.”Ben Morris blog.
  • 27. 27 Culture shapes infrastructure; infrastructure shapes culture: does your infrastructure support reuse? Data scientists working in government departments, making models Professionals within a data federation/ecosystem using trained model Exploration •Data Wrangling •DS Lab •Model scripting (untrained models) •Testing, Training, Model Evaluation •Version Control •Dependency Management Automate • Software unit tests • Model Training • Storage of trained models • Model Evaluation • Model Business Approval/Report Creation • Comparison vs current Live model (Champion/Challenger) Consume •Real-time model scoring engines •Automatic deployment of trained model artefacts •Dashboards and forecasts updated using new models •Model performance monitoring •Model output logging Involving: Analysts, Data Scientists, Engineers, Dev Ops, Business Stakeholders © 2018 Teradata
  • 28. 28 Culture shapes infrastructure; infrastructure shapes culture Data Scientist develops model ❑ Models developed by data scientist using preferred, locally installed analytic tools Data Scientist prepares power point ❑ A data scientist prepares a Powerpoint presentation on the benefits of a new model ❑ Manually creating standard model performance metrics of model Model approval committee ❑ Model business approval committee meeting to: See presentation; Discuss benefits/risks; Approve/Reject model Model manually transferred to relevant production environment ❑ Once approved, model exported to an Excel file of weights ❑ Excel manually transcribed into format suitable for the Production environment running on organisation-specific systems Model goes live ❑ New model goes live in production under close supervision © 2018 Teradata
  • 29. 29 Culture shapes infrastructure; infrastructure shapes culture Data Scientist develops model ❑ Models developed in standardised Python or R Automated tests ❑ Model is tested using a validation data set ❑ Report with model performance metrics is automatically generated Peer approval ❑ Model performance report is available in a UI for approval ❑ Approving in the UI will trigger automatically the bundling of models into business use case Business approval ❑ The business impact of the new bundle of models is estimated and displayed in a UI for the business team to approve the bundle Model goes live ❑ New model gets deployed automatically with A/B testing capability to validate business impact ❑ The performance of the models in production are continuously monitored © 2018 Teradata
  • 30. 30 Traditional productivity tools assume highly individualised, localised ways of working. Culture shapes infrastructure; infrastructure shapes culture: do your tools support re-use?
  • 31. 31 How we balance investment in paying down technical debt against building new infrastructure is not a neutral decision. It directly contributes to whether or not we develop a healthy maintenance culture. How much extra work are we willing to do, to reuse and create reusable data?
  • 33. 33 • Build stronger links between design and policy, using data Culture shapes infrastructure, infrastructure shapes culture
  • 34. 34 Policy is, still, too often made in an evidence free vacuum. Good data is one of the ways we help shine a light on policy gaps.
  • 35. 35 • Build stronger links between design and policy, using data • Apply serious effort to incorporating responsible design of the data lifecycle - data discovery, data collection, metadata generation - into design sprints Culture shapes infrastructure, infrastructure shapes culture
  • 36. 36 We have to be aware of the challenges and risks of indiscriminate data gathering. And acknowledge the power dynamics at play and inherent biases in data collection. Thinking beyond the ‘happy path’
  • 37. 37
  • 38. 38 • Build stronger links between design and policy, using data • Apply serious effort to incorporating responsible design of the data lifecycle - data discovery, data collection, metadata generation - into design sprints • Standardise and reward data reuse (both supply and demand) Culture shapes infrastructure, infrastructure shapes culture (part I)
  • 39. 39 • Think more critically about whether our tools and working environment (e.g. Google Analytics) broaden our view of the data ecosystem or narrows it • Raise the threshold of ‘good enough’ data quality (defining it would be a good start!) • Build a strong data commons by building a wide community of users Culture shapes infrastructure, infrastructure shapes culture (part II)
  • 40. 40 • Let’s think about the stories we tell - what are we emphasising and celebrating? Culture shapes infrastructure, infrastructure shapes culture (part III)
  • 42. 42 • We’re waking up to the brittleness of our infrastructure. • We’ve been here before. There’s infrastructure (social and software) from the digital transformation era we can leverage. • Efforts to build community and maintain a digital commons persist. Why I remain hopeful about the future of government infrastructure
  • 43. 43 Thank you. ©2018 Teradata Thank you. ©2018 Teradata