Executive Whispering for Linked Data Knud Möller Data Engineering http://kasabi.com
Nov 01, 2014
Executive Whispering for Linked Data
Knud MöllerData Engineeringhttp://kasabi.com
Executive Whispering for Linked Data
Knud MöllerData Engineeringhttp://kasabi.com
*
* Jarred McGinnis (Press Association)
Kasabi & Talis Consulting
Not yet standard procedure...
‘Cur
ious
Gir
l’ by
zub
row
on
Flic
kr (
CC
BY-
NC
2.0
)
What’s the goal?“Hi, we want to do
something with Linked Data!”
“Great, what’s the something?”
“Erhm, we’re not quite sure...”
‘Sce
ptic
al O
wl’
by F
rans
Badg
er o
n Fl
ickr
(C
C B
Y-N
C 2
.0)
Don’t start with the technology!
Start with a concrete problem:
• Lots of untapped potential in our data!
• Our search engine results aren’t so great.
• Our internal knowledge management isn’t working.
• Cost savings, productivity increase, ...
• (Everybody else is doing it...)
Common Issues
• “But it’s too hard!”
• “We don’t have the right skills!”
• “We don’t have the right infrastructure!”
• Data just isn’t sexy...
But it’s too hard!• actively dissuade people from doing too big
a project
• instead, go step by step:
• do a small project first, gain experience, show benefits
• then build on that
‘Ste
ps’ b
y ne
t_ef
ekt
on F
lickr
(C
C B
Y-N
C 2
.0)
”From Idea to Web -Linked Data Apps by Example”Benjamin Nowack, 14:00
But do we have the right skills?
• Probably not, but...
• no need to train everyone, build a small, diverse team
• get someone external in!
Kasabi Network:
But do we have the right infrastructure?• Probably not, but...
• you can outsource - LD can now be done as SaaS
• hosting, integrating, monetizing:
Kasabi“We decided to let Talis
take care of the hosting and
serving, so all we had to do
was worry about making
the data available.”
John Goodwin, Research Scientist,
Ordnance Survey
Data isn’t sexy...Data isn’t sexy...
... but pretty pictures are.
@kasabi
http://blog.kasabi.com
... but pretty pictures are.
... but pretty pictures are.
Take home• Don’t talk about technology - talk about
concrete problems and benefits.
• Don’t do everything at once - start small and build from there.
• No need to retrain everyone - build up expertise bit by bit, get external help.
• No need to invest in infrastructure - use SaaS options.
• Make demos, visualisations, mockups!
Links• Jarred McGinnis, Press Association. “Semantic News”:
http://www.slideshare.net/TalisConsulting/semantic-news
• Kasabi Data Marketplace:
http://kasabi.com/
• Kasabi News:
http://blog.kasabi.com
@kasabi
• Talis Consulting Case Studies:
http://consulting.talis.com/case-studies/
• Data Liberate:
http://dataliberate.com/
• Epimorphics:
http://www.epimorphics.com
Links, ctd.• Linking Open Data cloud diagram, by Richard Cyganiak and
Anja Jentzsch:
http://lod-cloud.net/
• ‘Curious Girl’ by zubrow on Flickr (CC BY-NC 2.0):
http://www.flickr.com/photos/zubrow/5688491904/
• ‘Sceptical Owl’ by FransBadger on Flickr (CC BY-NC 2.0):
http://www.flickr.com/photos/fransmayra/226264535/
• ‘Steps’ by net_efekt on Flickr (CC BY-NC 2.0):
http://www.flickr.com/photos/wheatfields/504844510/
• These slides:
http://www.slideshare.net/dunken69/executive-whispering-for-linked-data-11441808
These slides are licensed under Creative Commons CC BY-SA.