March 13, 2025 • 15 mins
In this episode, Richard Robinson chats with OMG Chairman & CEO Bill Hoffman about his new book, "Cracking the Data Code: Pragmatics for Better Management & Governance."
If you are interested in data, and why understanding language and jargon can help you crack the data code, this book is for you. If you’ve had a data challenge and have struggled to find a way to understand it, the practical foundational principles inside can help you frame your problem in a different way. And in doing so, help you crack the data code.
Rich is the Chief Strategist, Open Symbology & Standards at Bloomberg; Co-Chairof the OMG's Financial Sector Domain Task Force, and an OMG Board Member.
Visit the OMG Standards Development Organization at https://omg.org
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:09):
Hello, I'm Karen Quatromoni,
Director of Public Relations forObject Management Group, OMG,
and welcome to our OMGPodcast series. At OMG,
we're known for driving industrystandards in building tech communities.
Today we're here with OMG CEO and ChairmanBill Hoffman who will lead today's
podcast session. Bill?
(00:29):
Thanks Karen. Hey, we're heretoday with Rich Robinson. Rich,
do you want to spend a coupleminutes to introduce yourself.
Sure, and thanks for havingme, Bill. So Rich Robinson,
I am a Chief Strategistfor Open Symbology and
Standards at Bloomberg.
Also for the Object Management Group,
(00:50):
I am co-chair of the financialsector DTF and on the
Board.
Great, thanks. And Rich, you'vebeen around OMG for quite a while.
Talk about your new book,
"Cracking the Data Code Pragmatics forBetter Management and Governance." Can
you give me a little bit ofoverview of what you wrote?
Sure.
(01:12):
So it's a follow up to aprevious book that I wrote in
2021, also published byBusiness Expert Press.
So in the first book I introducedthe idea of using applied
linguistic tools,
specifically concepts around communitiesof practice within data management.
So the new book goesfurther into that into,
(01:36):
so the thing is that there's allthese established tools within Applied
Linguistics and data managersand technologists today just
dunno about or consider
and that they couldreally use to leverage.
They have concepts like semantics andontologies are taken from linguistics,
but they don't build out pastthat within data management.
(02:01):
The thing is, it's not overly written,
it's not written overlytechnical or complex.
It's written for the layman,
but also for well-established datatechnology practices, practitioners.
I've been in the industryfor 30 plus years.
I won't say the exact number,
(02:22):
but so I have experience across front,
mid back office and financial services,both under operators, technology side
side.
And I've seen this happen withinall the different areas I've been
in dealing with regulators,dealing with technology.
(02:43):
And what really comes down to isunderstanding the biases and context
that exists within allthe things that we do.
And everybody comes from a differentcommunity and has a different perspective.
So within this book, Iexplore more into the
(03:04):
use of context bias and thehistory of data storage to
ml AI use of LLMs,
NLP and all the new topics that we're
talking about.
But I do try and relate it back toeveryday examples and analogies to make
it easier,
(03:25):
easier reading as opposed tobeing this textbook. Excellent.
Excellent, excellent. I'll tellyou, I think it's very timely.
What questions did you see in themarket that you needed to address?
So as I mentioned,
there's no real inclusion of appliedlinguistics tools and methods within data
management today. There'stalks of semantics in context,
(03:49):
but as individual conceptson their own right,
someone else says, oh,that's just semantics. Oh,
well we need to talk about context.
But they're really components of alarger system that they belong to.
Within linguistics and useof semantics in context
is individual things withindata management and how we talk about data and even
(04:12):
ontologies and whatnot.
It really handicaps theapplication of those concepts,
but also means they'remisapplied. A lot of times.
Pragmatics is one of the things thathelps bring semantics and context
together.
It's sort of like an overarchinggrouping of a lot of linguistic things
together. So what's important about it?
(04:37):
Consider that the world ofdata isn't just about data.
It's kind of useless unless it's beingcommunicated. So really data is all about
enabling communication in human discourse.
So our conversation,
it's two people talking togetherand this is where a miscommunication
misunderstanding happens.It's that communication bit.
(05:00):
So does any wonder that this is alsothe biggest problem in data as well?
It's not the data, it'swhen the data gets shared.
So simplistically,
you can think of pragmatics as not justgetting a handle on the context of a
communication and the meaningof the words or the data,
but also the history that surroundsthat communication. Why is it happening?
(05:23):
The understanding of the biases andintent of both the sending party and the
receiving party. So pragmaticsencompasses all of this
in human communication.
We easily repair or accommodateor the linguistic terms for this.
That's what it means. Basically,if we don't understand something,
we ask for clarification or we take intoaccount someone's for accommodation,
(05:47):
we take into account someone speaksqueen's English versus more American
English.
So if you're speakingto someone in the UK,
they say jumper, that meanssweater. That's not done in data.
We don't even capture thosekinds of concepts and data.
We trust in the data andstart from this false
(06:09):
premise that the data we send hassome how been uniquely defined
so well that it's always going to beunderstood by the people that receive it
and that's just not realistic
I think.
So something that I've dealt withrecently is the concept of entity.
(06:32):
I've dated people,
talked about entities with peopleinvolved with corporate structures.
So they're building outdata models and ontologies
halfway through the conversation.
So it's these technologistssitting with business people.
Basically halfwaythrough the conversation,
the business people were lostbecause they assume the word entity
(06:54):
was referring to a companybeing that they were just using
entity as another word to saycompany. But in data speak,
an entity is just a genericterm for a data object.
So our expertise and where we comefrom historically and everything else
tends to biases to lead,
(07:15):
to assume that everybody else knows whatwe're saying when we say something and
they assume the sameinterpretations when we say it.
So without a pragmaticsview, governing that data,
conversation errors aremuch more likely to occur.
So that's where I start at
the beginning of the book and that'swhere the problems I saw and that's what
(07:38):
I'm trying to address.
Excellent, excellent. That's fascinating.
So we obviously build standardsat the Object Management Group.
How are these topics related tothe standards work that's going on?
So I think standards is a great thing.
I belong to group and I belong to anumber of other standards organizations
around the globe, but standardsare also very misunderstood.
(08:03):
So many think of standards as theseoverarching rules that make everything the
same in regard to any other influences.So if it's a standard, it's a standard.
But I'm a fan of Tannenbaum who said thatthe greatest thing about standards is
that there are so many, and Idon't see this as a bad thing.
A lot of people use it as a derisivething, but I don't see it as a bad thing.
(08:26):
This brings us back to language.
So data being an instance of languageis bound by the same rules as language,
and that is, it evolves and changesand what it evolves and changes.
Variations are sort of bound bythe specific community of practice
that uses that language. Iknow I'm saying a lot of words.
(08:47):
So look at medicine. Soit's all medical stuff.
So they must use the sameterms and definitions.
You just assume that it'sall medical, but they don't.
Pediatricians and oncologists are goingto have their own jargon and dialects,
much like Brazilian,Portuguese and European.
Portuguese pediatrician isn't goingto walk in a room of oncologists
(09:11):
talking oncologists jargon and immediatelyunderstand what they're talking
about.
It's going to take some interpretationand figuring out for them as opposed to
the oncologists just nativelyknow what they're saying.
So is the answer to force general overall
standardization?
I'd argue on the side that youstandardize within communities of
(09:34):
practice where there'll be asignificantly less churn and
change and then focus on the translatingacross those different communities.
So we're talking about standards.
The divergence of human language wasn'tsaw by trying to introduce Esperanto.
Esperanto is meant tostandardize human language,
(09:55):
but it's much more effective totranslate between English and Spanish and
Japanese and also to translate betweenwhat queen's English calls a jumper and
American calls a sweaterbecause standards within
communities create deficiencies,
but mainly where they canbe applied specifically.
So the few effective broadoverarching standards that exist,
(10:18):
say like XML,
only become useful and effective whenthey're applied to specific use cases.
They're structural, they'renot specific standards, right?
The same because at the same time,
two specific XML implementationstypically are not compatible with
each other and not interoperable at all.
(10:42):
So I think for OMG
participants, they understand this.
If you look at corbo, which isa foundational thing, right?
Primary focus is on translation.
While the various groups developingstandards that OMG are designed
based on communities of practice,
(11:03):
bringing together expertsthat share the same processes,
they share the same goals and languagethat create specific use case driven
solutions.
Excellent.
You want to talk a little bit about yourorganization and why you guys joined?
The Object Management Group? It'sbeen members for several years.
Over 10.
(11:24):
I think Bloomberg is small littledata company based out of New York.
We provide services to thefinancial services community
mainly, but also to other communities,
whether that's governmentalor life sciences and so
(11:44):
on. Corporate desks ingeneral and on and on.
As I mentioned, OMG is a great place for
looking at standards,
especially from a technical leveland understanding the things that
run around it technically.
(12:05):
So when we were lookingat the issues around
financial instrument identification,
understanding that it wasn't justidentifying a financial instrument,
but actually looking at a financialinstrument in different contexts.
There's that word again, butalso through different use cases.
So looking at more of that pragmaticsview of how a financial instrument might
(12:29):
be treated throughout anentire lifecycle that has or
through different interactions.
So like a trader that needs to seeprices across multiple exchanges
versus someone who is justsettling in one country location
versus a risk person whomight want to look at exposure
(12:49):
across multiplejurisdictions or currencies.
The expertise of Object Managementgroup, I think lent to that,
and that's what I'm describingobviously is what came out to the
financial instrument,
global identifier or the FIGI (FinancialInstrument Global Identifier).
Well, that's been a big success.No doubt I've been out there,
(13:10):
but there's millions andmillions of downloads.
So
I think in the past six months we've had
250 billion requests
for the FIGI data and there's
1.6 billion FIGIs that havebeen issued over the past 10 to
(13:33):
12 year lifespan.
Yep. It's been quite the success.
And we're in the middleof revision 1.3. So.
There you go. There you go. Well,it's going to continue to move ahead.
Has OMG membership helped Bloomberg?
So I would say OMG is critical togetting the subject matter expertise
(13:57):
to evaluate and enhanceFIGI, which Bloomberg uses,
obviously it's object managementgroup standards, open data standard,
anybody can use it.
Bloomberg has seen thatit's good as a foundational
skeleton to build our financial instrument
(14:19):
reference data around. So
Object Management Group being a standardsdevelopment organization that supports
open data and openstandards is a great thing.
And so from that perspective,
the relationship that Bloomberg has atObject Management Group and expertise
(14:42):
that we're exposed to is fantastic.
That's great. Super. That's great.Rich, I appreciate your time today.
Any final thoughts youwant to leave us with?
Please do get the book"Cracking the Data Code."
It's available anywhere, right? So online.
(15:03):
Your favorite online retailer acrossthe globe. So not just on Amazon,
but Barnes and Noble andalso in the UK. I think it's
HMS.
There's a number of onlineretailers in Europe and in Asia.
So it's accessible all throughthere. Also through O'Reilly,
if you have an O'Reilly subscription,if you get the hard copy,
(15:26):
bring it to me. I'llsign it. There you go.
I'll buy you a beverage of your choosing.
So the name of the book is"Cracking the Data Code:
Pragmatics for Better Managementand Governance." Rich,
thank you so much for your time today.
Thank you very much, Bill for having me.
Hello, I'm Karen Quatromoni,
Director of Public Relations forObject Management Group, OMG,
and welcome to our OMGPodcast series. At OMG,
we're known for driving industrystandards in building tech communities.
Today we're here with OMG CEO and ChairmanBill Hoffman who will lead today's
podcast session. Bill?
(00:29):
Thanks Karen. Hey, we're heretoday with Rich Robinson. Rich,
do you want to spend a coupleminutes to introduce yourself.
Sure, and thanks for havingme, Bill. So Rich Robinson,
I am a Chief Strategistfor Open Symbology and
Standards at Bloomberg.
Also for the Object Management Group,
(00:50):
I am co-chair of the financialsector DTF and on the
Board.
Great, thanks. And Rich, you'vebeen around OMG for quite a while.
Talk about your new book,
"Cracking the Data Code Pragmatics forBetter Management and Governance." Can
you give me a little bit ofoverview of what you wrote?
Sure.
(01:12):
So it's a follow up to aprevious book that I wrote in
2021, also published byBusiness Expert Press.
So in the first book I introducedthe idea of using applied
linguistic tools,
specifically concepts around communitiesof practice within data management.
So the new book goesfurther into that into,
(01:36):
so the thing is that there's allthese established tools within Applied
Linguistics and data managersand technologists today just
dunno about or consider
and that they couldreally use to leverage.
They have concepts like semantics andontologies are taken from linguistics,
but they don't build out pastthat within data management.
(02:01):
The thing is, it's not overly written,
it's not written overlytechnical or complex.
It's written for the layman,
but also for well-established datatechnology practices, practitioners.
I've been in the industryfor 30 plus years.
I won't say the exact number,
(02:22):
but so I have experience across front,
mid back office and financial services,both under operators, technology side
side.
And I've seen this happen withinall the different areas I've been
in dealing with regulators,dealing with technology.
(02:43):
And what really comes down to isunderstanding the biases and context
that exists within allthe things that we do.
And everybody comes from a differentcommunity and has a different perspective.
So within this book, Iexplore more into the
(03:04):
use of context bias and thehistory of data storage to
ml AI use of LLMs,
NLP and all the new topics that we're
talking about.
But I do try and relate it back toeveryday examples and analogies to make
it easier,
(03:25):
easier reading as opposed tobeing this textbook. Excellent.
Excellent, excellent. I'll tellyou, I think it's very timely.
What questions did you see in themarket that you needed to address?
So as I mentioned,
there's no real inclusion of appliedlinguistics tools and methods within data
management today. There'stalks of semantics in context,
(03:49):
but as individual conceptson their own right,
someone else says, oh,that's just semantics. Oh,
well we need to talk about context.
But they're really components of alarger system that they belong to.
Within linguistics and useof semantics in context
is individual things withindata management and how we talk about data and even
(04:12):
ontologies and whatnot.
It really handicaps theapplication of those concepts,
but also means they'remisapplied. A lot of times.
Pragmatics is one of the things thathelps bring semantics and context
together.
It's sort of like an overarchinggrouping of a lot of linguistic things
together. So what's important about it?
(04:37):
Consider that the world ofdata isn't just about data.
It's kind of useless unless it's beingcommunicated. So really data is all about
enabling communication in human discourse.
So our conversation,
it's two people talking togetherand this is where a miscommunication
misunderstanding happens.It's that communication bit.
(05:00):
So does any wonder that this is alsothe biggest problem in data as well?
It's not the data, it'swhen the data gets shared.
So simplistically,
you can think of pragmatics as not justgetting a handle on the context of a
communication and the meaningof the words or the data,
but also the history that surroundsthat communication. Why is it happening?
(05:23):
The understanding of the biases andintent of both the sending party and the
receiving party. So pragmaticsencompasses all of this
in human communication.
We easily repair or accommodateor the linguistic terms for this.
That's what it means. Basically,if we don't understand something,
we ask for clarification or we take intoaccount someone's for accommodation,
(05:47):
we take into account someone speaksqueen's English versus more American
English.
So if you're speakingto someone in the UK,
they say jumper, that meanssweater. That's not done in data.
We don't even capture thosekinds of concepts and data.
We trust in the data andstart from this false
(06:09):
premise that the data we send hassome how been uniquely defined
so well that it's always going to beunderstood by the people that receive it
and that's just not realistic
I think.
So something that I've dealt withrecently is the concept of entity.
(06:32):
I've dated people,
talked about entities with peopleinvolved with corporate structures.
So they're building outdata models and ontologies
halfway through the conversation.
So it's these technologistssitting with business people.
Basically halfwaythrough the conversation,
the business people were lostbecause they assume the word entity
(06:54):
was referring to a companybeing that they were just using
entity as another word to saycompany. But in data speak,
an entity is just a genericterm for a data object.
So our expertise and where we comefrom historically and everything else
tends to biases to lead,
(07:15):
to assume that everybody else knows whatwe're saying when we say something and
they assume the sameinterpretations when we say it.
So without a pragmaticsview, governing that data,
conversation errors aremuch more likely to occur.
So that's where I start at
the beginning of the book and that'swhere the problems I saw and that's what
(07:38):
I'm trying to address.
Excellent, excellent. That's fascinating.
So we obviously build standardsat the Object Management Group.
How are these topics related tothe standards work that's going on?
So I think standards is a great thing.
I belong to group and I belong to anumber of other standards organizations
around the globe, but standardsare also very misunderstood.
(08:03):
So many think of standards as theseoverarching rules that make everything the
same in regard to any other influences.So if it's a standard, it's a standard.
But I'm a fan of Tannenbaum who said thatthe greatest thing about standards is
that there are so many, and Idon't see this as a bad thing.
A lot of people use it as a derisivething, but I don't see it as a bad thing.
(08:26):
This brings us back to language.
So data being an instance of languageis bound by the same rules as language,
and that is, it evolves and changesand what it evolves and changes.
Variations are sort of bound bythe specific community of practice
that uses that language. Iknow I'm saying a lot of words.
(08:47):
So look at medicine. Soit's all medical stuff.
So they must use the sameterms and definitions.
You just assume that it'sall medical, but they don't.
Pediatricians and oncologists are goingto have their own jargon and dialects,
much like Brazilian,Portuguese and European.
Portuguese pediatrician isn't goingto walk in a room of oncologists
(09:11):
talking oncologists jargon and immediatelyunderstand what they're talking
about.
It's going to take some interpretationand figuring out for them as opposed to
the oncologists just nativelyknow what they're saying.
So is the answer to force general overall
standardization?
I'd argue on the side that youstandardize within communities of
(09:34):
practice where there'll be asignificantly less churn and
change and then focus on the translatingacross those different communities.
So we're talking about standards.
The divergence of human language wasn'tsaw by trying to introduce Esperanto.
Esperanto is meant tostandardize human language,
(09:55):
but it's much more effective totranslate between English and Spanish and
Japanese and also to translate betweenwhat queen's English calls a jumper and
American calls a sweaterbecause standards within
communities create deficiencies,
but mainly where they canbe applied specifically.
So the few effective broadoverarching standards that exist,
(10:18):
say like XML,
only become useful and effective whenthey're applied to specific use cases.
They're structural, they'renot specific standards, right?
The same because at the same time,
two specific XML implementationstypically are not compatible with
each other and not interoperable at all.
(10:42):
So I think for OMG
participants, they understand this.
If you look at corbo, which isa foundational thing, right?
Primary focus is on translation.
While the various groups developingstandards that OMG are designed
based on communities of practice,
(11:03):
bringing together expertsthat share the same processes,
they share the same goals and languagethat create specific use case driven
solutions.
Excellent.
You want to talk a little bit about yourorganization and why you guys joined?
The Object Management Group? It'sbeen members for several years.
Over 10.
(11:24):
I think Bloomberg is small littledata company based out of New York.
We provide services to thefinancial services community
mainly, but also to other communities,
whether that's governmentalor life sciences and so
(11:44):
on. Corporate desks ingeneral and on and on.
As I mentioned, OMG is a great place for
looking at standards,
especially from a technical leveland understanding the things that
run around it technically.
(12:05):
So when we were lookingat the issues around
financial instrument identification,
understanding that it wasn't justidentifying a financial instrument,
but actually looking at a financialinstrument in different contexts.
There's that word again, butalso through different use cases.
So looking at more of that pragmaticsview of how a financial instrument might
(12:29):
be treated throughout anentire lifecycle that has or
through different interactions.
So like a trader that needs to seeprices across multiple exchanges
versus someone who is justsettling in one country location
versus a risk person whomight want to look at exposure
(12:49):
across multiplejurisdictions or currencies.
The expertise of Object Managementgroup, I think lent to that,
and that's what I'm describingobviously is what came out to the
financial instrument,
global identifier or the FIGI (FinancialInstrument Global Identifier).
Well, that's been a big success.No doubt I've been out there,
(13:10):
but there's millions andmillions of downloads.
So
I think in the past six months we've had
250 billion requests
for the FIGI data and there's
1.6 billion FIGIs that havebeen issued over the past 10 to
(13:33):
12 year lifespan.
Yep. It's been quite the success.
And we're in the middleof revision 1.3. So.
There you go. There you go. Well,it's going to continue to move ahead.
Has OMG membership helped Bloomberg?
So I would say OMG is critical togetting the subject matter expertise
(13:57):
to evaluate and enhanceFIGI, which Bloomberg uses,
obviously it's object managementgroup standards, open data standard,
anybody can use it.
Bloomberg has seen thatit's good as a foundational
skeleton to build our financial instrument
(14:19):
reference data around. So
Object Management Group being a standardsdevelopment organization that supports
open data and openstandards is a great thing.
And so from that perspective,
the relationship that Bloomberg has atObject Management Group and expertise
(14:42):
that we're exposed to is fantastic.
That's great. Super. That's great.Rich, I appreciate your time today.
Any final thoughts youwant to leave us with?
Please do get the book"Cracking the Data Code."
It's available anywhere, right? So online.
(15:03):
Your favorite online retailer acrossthe globe. So not just on Amazon,
but Barnes and Noble andalso in the UK. I think it's
HMS.
There's a number of onlineretailers in Europe and in Asia.
So it's accessible all throughthere. Also through O'Reilly,
if you have an O'Reilly subscription,if you get the hard copy,
(15:26):
bring it to me. I'llsign it. There you go.
I'll buy you a beverage of your choosing.
So the name of the book is"Cracking the Data Code:
Pragmatics for Better Managementand Governance." Rich,
thank you so much for your time today.
Thank you very much, Bill for having me.
Popular Podcasts
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
Therapy Gecko
An unlicensed lizard psychologist travels the universe talking to strangers about absolutely nothing. TO CALL THE GECKO: follow me on https://www.twitch.tv/lyleforever to get a notification for when I am taking calls. I am usually live Mondays, Wednesdays, and Fridays but lately a lot of other times too. I am a gecko.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.