ShipTalk S03E07 - Nic Acton talks with Ron Gidron (CEO, xtype.io)

00:47:00
https://www.youtube.com/watch?v=6FcRg7CoJRs

Resumen

TLDRIn this episode of Ship Talk, Nick Acton interviews Ron Gon from XType, who discusses the complexities of DevOps within the ServiceNow platform. Ron shares his extensive experience in software delivery and automation, emphasizing the importance of multi-instance management to maximize ServiceNow investments. The conversation highlights the differences between traditional DevOps and ServiceNow DevOps, focusing on the need for visibility, governance, and automation. Ron also addresses the challenges organizations face when integrating various systems and data, and the evolving role of AI in streamlining development processes. He encourages organizations to avoid reinventing the wheel and to consider the complexities of their environments when adopting new platforms.

Para llevar

  • 🚀 XType helps maximize ServiceNow investments through multi-instance management.
  • 🔍 Visibility, governance, and automation are key capabilities for effective DevOps.
  • ⚙️ ServiceNow DevOps differs from traditional DevOps due to vendor and custom code alignment.
  • 💡 Organizations should avoid reinventing the wheel and leverage existing platforms.
  • 📊 Inconsistencies between environments can complicate deployments.
  • 🤖 AI can enhance the software lifecycle by predicting testing needs and automating decisions.
  • 📜 Governance ensures compliance and consistency across development processes.
  • 🔧 Version control is an underrated tool in the DevOps space.
  • 💻 JavaScript is a preferred language for quick automation tasks.
  • 🚫 The buzzword 'agile' is often overused and could be retired.

Cronología

  • 00:00:00 - 00:05:00

    The video features a conversation between Nick Acton and Ron Gon, discussing the challenges and insights related to DevOps, particularly in the context of ServiceNow. Ron shares his extensive background in software delivery and automation, highlighting his experience in the DevOps space since the 90s.

  • 00:05:00 - 00:10:00

    Ron explains that XType is a multi-instance management platform for ServiceNow, aimed at helping customers maximize their investment through visibility, governance, and automation. He emphasizes the differences between traditional development and platform-based development, particularly in how codebases are managed.

  • 00:10:00 - 00:15:00

    The discussion shifts to the unique aspects of ServiceNow DevOps compared to traditional DevOps. Ron points out that when using a platform, developers must work within the constraints of the vendor's codebase, which can lead to challenges in maintaining alignment between custom code and vendor updates.

  • 00:15:00 - 00:20:00

    Nick and Ron explore the decision-making process for organizations considering whether to build software from scratch or utilize existing platforms like ServiceNow. Ron advises against reinventing the wheel unless absolutely necessary, emphasizing the importance of cost of ownership and technical debt.

  • 00:20:00 - 00:25:00

    Ron discusses the importance of understanding that systems do not live in isolation, and organizations should consider future integrations and the broader ecosystem when making software decisions. He suggests that starting from scratch may not be practical for most enterprise use cases.

  • 00:25:00 - 00:30:00

    The conversation touches on the challenges organizations face when adopting ServiceNow and the importance of integrating DevOps practices. Ron highlights the need for a structured approach to development and deployment, including the use of CI/CD processes.

  • 00:30:00 - 00:35:00

    Ron shares insights on the tools available for ServiceNow development, including workflow studios and APIs, and emphasizes the importance of managing code changes across different environments to ensure consistency and reliability.

  • 00:35:00 - 00:40:00

    The discussion highlights the significance of visibility, automation, and governance in managing DevOps processes effectively. Ron explains how organizations can leverage these capabilities to improve their development workflows and maintain compliance.

  • 00:40:00 - 00:47:00

    Finally, Ron reflects on the ongoing challenges in the DevOps space, emphasizing that while technology and methodologies evolve, the fundamental complexities of managing software development and operations remain consistent. He expresses optimism about the future of DevOps, particularly with the integration of AI and automation to enhance decision-making and streamline processes.

Ver más

Mapa mental

Vídeo de preguntas y respuestas

  • What is XType's mission?

    XType aims to help ServiceNow customers maximize their investment through multi-instance management.

  • What are the main capabilities of XType?

    The main capabilities are visibility, governance, and automation.

  • How does ServiceNow DevOps differ from traditional DevOps?

    ServiceNow DevOps involves working within a platform with vendor code and custom code, requiring alignment between the two.

  • What should organizations consider when adopting ServiceNow?

    Organizations should evaluate whether to build from scratch or use existing platforms like ServiceNow based on their specific needs.

  • What are the challenges in managing multiple environments?

    Inconsistencies between environments, such as data differences and plugin versions, can complicate deployments.

  • What is the role of AI in DevOps?

    AI can predict testing needs, automate decision-making, and enhance the software lifecycle.

  • What is the importance of governance in DevOps?

    Governance ensures compliance and consistency across development processes and environments.

  • What is the most underrated tool in DevOps?

    Version control is often underrated but is crucial for managing code.

  • What programming language does Ron prefer for automation?

    Ron prefers JavaScript for quick automation tasks.

  • What buzzword does Ron wish would go away?

    Ron wishes the buzzword 'agile' would go away.

Ver más resúmenes de vídeos

Obtén acceso instantáneo a resúmenes gratuitos de vídeos de YouTube gracias a la IA.
Subtítulos
en
Desplazamiento automático:
  • 00:00:03
    [Music]
  • 00:00:09
    hey everybody Welcome to ship talk my
  • 00:00:11
    name is Nick Acton I'm a sales engineer
  • 00:00:13
    here at herness I'm here with Ron gon
  • 00:00:15
    from xtype which is doing a lot of
  • 00:00:18
    devops for other modalities that we all
  • 00:00:21
    may not be as familiar with with service
  • 00:00:23
    now um and you know we're going to talk
  • 00:00:25
    a little bit about some of the
  • 00:00:26
    challenges and everything there but uh
  • 00:00:28
    Ron thank you so much for coming on the
  • 00:00:30
    show wanted to give you an opportunity
  • 00:00:31
    to um tell us a little bit about
  • 00:00:33
    yourself and then let's take the
  • 00:00:34
    conversation from
  • 00:00:36
    there yeah thanks Nick uh great to be
  • 00:00:39
    here I appreciate it you also pronounced
  • 00:00:40
    my name right which is uh which is a
  • 00:00:43
    very rare so thanks it's great to be
  • 00:00:46
    here thank you so let's uh let's get
  • 00:00:50
    started and of course we'll we'll make
  • 00:00:51
    this more conversational as we go but um
  • 00:00:54
    you know definitely want to learn more
  • 00:00:55
    about you in general what what is your
  • 00:00:58
    background and how did you get in into
  • 00:01:00
    software delivery devops everything like
  • 00:01:03
    that great so um I'm Israel born in Bren
  • 00:01:07
    I started my career uh my Tech Career in
  • 00:01:10
    the 90s uh um and I went to work for a
  • 00:01:13
    test automation company so automation
  • 00:01:15
    was kind of right there in the beginning
  • 00:01:18
    I worked for one of the pioneers of the
  • 00:01:19
    test uh test automation industry this
  • 00:01:22
    company called Mercury
  • 00:01:24
    interactive um and we built load and
  • 00:01:27
    functional testing tools automated
  • 00:01:29
    testing tools
  • 00:01:30
    so I spent 10 years there kind of saw
  • 00:01:33
    the internet you know come alive and um
  • 00:01:36
    then spent another 10 years in um two
  • 00:01:38
    different monitoring companies database
  • 00:01:42
    monitoring transaction tracking things
  • 00:01:45
    of that nature and then went back into
  • 00:01:47
    the automation space uh round about uh
  • 00:01:51
    2008 I worked for a one of the
  • 00:01:55
    early devops I think uh startups that
  • 00:01:59
    were there compan noio did release
  • 00:02:00
    automation for you know uh on Prem Java
  • 00:02:04
    servers and stuff like that and went on
  • 00:02:06
    to different company called the uc4
  • 00:02:10
    atomic helped build a Again release
  • 00:02:13
    automation develops product and um you
  • 00:02:15
    know so I've I've been in the space for
  • 00:02:17
    a long time automation devops are
  • 00:02:20
    something that I've kind of done my
  • 00:02:22
    whole
  • 00:02:23
    career nice so you've been there since
  • 00:02:25
    really the really the beginning of of
  • 00:02:27
    devops and um probably have an interest
  • 00:02:30
    perspective I've only been in it for
  • 00:02:31
    coming up on a decade so um you know
  • 00:02:34
    you're going to have a lot more insight
  • 00:02:36
    there but um you know what what is it
  • 00:02:39
    you're all doing today and maybe we'll
  • 00:02:41
    work backwards uh what are you all doing
  • 00:02:43
    today with xtype and and why is that
  • 00:02:45
    important all right great so X type is a
  • 00:02:49
    multi-instance management platform for
  • 00:02:51
    service now and our mission is to help
  • 00:02:55
    um service now customers maximize their
  • 00:02:57
    investment in service now through
  • 00:03:00
    through um multi-instance management
  • 00:03:02
    which I'll get into in a minute uh three
  • 00:03:05
    main capabilities that you know might is
  • 00:03:07
    going to is going to come around to
  • 00:03:08
    devops in a minute that's visibility
  • 00:03:11
    governance and
  • 00:03:13
    automation um in order to
  • 00:03:18
    understand how what I've just said is
  • 00:03:20
    connected to devops a good place to
  • 00:03:22
    start is by realizing that um a lot of
  • 00:03:27
    platforms out there service now in
  • 00:03:29
    particular don't exactly work the same
  • 00:03:32
    way that you know um generalized
  • 00:03:35
    development works I mean uh if you look
  • 00:03:37
    at service now development you're going
  • 00:03:38
    to see the exact you know you're going
  • 00:03:40
    to see JavaScript and you're going to
  • 00:03:41
    see a lot of the same a lot of the same
  • 00:03:43
    programming languages however the work
  • 00:03:46
    is being done whether you're developing
  • 00:03:48
    an app from scratch or customizing an
  • 00:03:50
    existing app is done within a within a a
  • 00:03:54
    platform um so you can think of it I if
  • 00:03:57
    you want to go back a little bit back in
  • 00:03:58
    history you can think of um you know
  • 00:04:01
    other platforms that were there sap was
  • 00:04:03
    there and so if you're going to write an
  • 00:04:05
    sap application you're doing it within
  • 00:04:07
    the boundaries of sap and you're
  • 00:04:09
    attached to that architecture today if
  • 00:04:11
    you're developing on the Salesforce
  • 00:04:13
    platform you're doing it in a particular
  • 00:04:15
    way there's multiple ways but you're
  • 00:04:16
    doing it on their architecture so I
  • 00:04:18
    think the main difference is if you're
  • 00:04:21
    in out in the open you work on software
  • 00:04:25
    projects that you know the
  • 00:04:28
    architecture uh is basically chosen by
  • 00:04:30
    the team programming language chosen by
  • 00:04:33
    the team databases you know
  • 00:04:35
    infrastructure everything's chosen by
  • 00:04:37
    the team um at the end you know modern
  • 00:04:40
    development all starts with a code base
  • 00:04:43
    maybe even your infrastructures as code
  • 00:04:46
    it all starts with a Version Control
  • 00:04:49
    with kit across across everything but
  • 00:04:53
    then there's a lot of development in the
  • 00:04:54
    world that's done on top of existing
  • 00:04:57
    platforms and so the base development
  • 00:05:00
    architecture is just not the
  • 00:05:02
    same right so I the service now devops
  • 00:05:06
    is is fascinating to me because and you
  • 00:05:09
    bring up a lot of excellent points which
  • 00:05:10
    is you know at the end of the day we're
  • 00:05:14
    all doing devops on a platform of some
  • 00:05:16
    kind whether that happens to be service
  • 00:05:18
    now or whether that happens to be a
  • 00:05:20
    virtual machine sitting in Amazon web
  • 00:05:22
    services running you know something a
  • 00:05:24
    little less you know no code and
  • 00:05:27
    everything is servic now would say
  • 00:05:28
    although there's plenty of code behind
  • 00:05:30
    service now um you know what's what
  • 00:05:33
    what's maybe some of the major
  • 00:05:35
    differentiations someone should be
  • 00:05:36
    thinking about when they're thinking of
  • 00:05:38
    like traditional devops versus like
  • 00:05:40
    service now devops or even broader SAS
  • 00:05:42
    devops for For What It's For What It's
  • 00:05:44
    Worth right so I I think this applies to
  • 00:05:46
    the brother SAS devops the the I think
  • 00:05:50
    the the main difference is is this um
  • 00:05:54
    when you choose a platform as your base
  • 00:05:57
    and there's plenty good reasons to do
  • 00:05:58
    that right certain type of applications
  • 00:06:00
    you wouldn't want to start from scratch
  • 00:06:02
    why would you so let me give you an
  • 00:06:03
    example uh you you you want a ticketing
  • 00:06:06
    system you want a system to manage it
  • 00:06:07
    you want a system to manage HR you want
  • 00:06:09
    a system to manage lots of things right
  • 00:06:12
    you can you could uh just build it from
  • 00:06:14
    scratch right we'll start from scratch
  • 00:06:16
    make deer and run from there but You' be
  • 00:06:20
    building so much of the stuff that's
  • 00:06:21
    already built that if you're an
  • 00:06:23
    organization just doesn't make any sense
  • 00:06:24
    to build that type of application
  • 00:06:26
    yourself so you're going to go out and
  • 00:06:27
    find a platform that already does it
  • 00:06:30
    let's say service now right now you're
  • 00:06:32
    going to you're so you're going to start
  • 00:06:33
    from some sort of a base and you're
  • 00:06:35
    going to customize it to your own needs
  • 00:06:37
    now you've got that platform you're
  • 00:06:38
    going to develop some new stuff on the
  • 00:06:40
    platform which is very uh very common
  • 00:06:43
    place and so like I said you you are
  • 00:06:46
    starting at a point that's different to
  • 00:06:49
    traditional development and so now you
  • 00:06:51
    have and and this is and this is the key
  • 00:06:54
    so now you have your codebase but you
  • 00:06:56
    also have the vendor codebase they're
  • 00:06:58
    two separate code basis because the
  • 00:07:00
    vendor is working on the platform right
  • 00:07:02
    service now releases every every six
  • 00:07:04
    months there's a new release sales
  • 00:07:05
    course every you know whatever there
  • 00:07:07
    there's a release so you can think of it
  • 00:07:09
    almost like two different code bases
  • 00:07:11
    that kind of have to align the whole
  • 00:07:13
    time your code and the vendor code right
  • 00:07:16
    you don't own the vendor code you have
  • 00:07:18
    no access to to their uh um repos or you
  • 00:07:24
    know codebase you own yours but you're
  • 00:07:27
    dependent on them because if they make
  • 00:07:29
    make a change to an API then your
  • 00:07:31
    changes have to happen and then there's
  • 00:07:33
    other stuff that is that is very
  • 00:07:35
    different because some sometimes you
  • 00:07:37
    know um you'll add a Plugin or a third
  • 00:07:41
    party application it's just it's just a
  • 00:07:43
    different mode of building building on a
  • 00:07:46
    platform versus building on you know uh
  • 00:07:50
    codebase right and you know it's it's
  • 00:07:53
    interesting to have these conversations
  • 00:07:54
    with customers and you know I've got
  • 00:07:56
    many that I've been writing down but um
  • 00:07:58
    you know I really happy to be talking
  • 00:08:00
    with you on the podcast just because
  • 00:08:02
    this is uh becoming more and more
  • 00:08:03
    prevalent across the space I think one
  • 00:08:05
    one really interesting question that I
  • 00:08:07
    I've kind of thought of and you I think
  • 00:08:09
    you'd have fantastic insight into is you
  • 00:08:12
    know we have a lot of customers that we
  • 00:08:13
    talk with at harness or prospects we
  • 00:08:15
    talk with at harness that are sometimes
  • 00:08:18
    relatively early or early in their
  • 00:08:20
    journey in terms of hey we are going to
  • 00:08:23
    utilize software in order to solve a
  • 00:08:25
    specific set of business use cases and
  • 00:08:28
    everything and one of those key early
  • 00:08:30
    questions they are kind of asking
  • 00:08:31
    themselves is the do we bring on a team
  • 00:08:34
    of full stack engineers and build it end
  • 00:08:36
    to end or do we go with something like a
  • 00:08:39
    Pega service now uh Salesforce or
  • 00:08:42
    anything like that uh you know how would
  • 00:08:46
    you maybe talk to someone through that
  • 00:08:49
    situation um what are some parameters
  • 00:08:51
    and everything they should consider to
  • 00:08:53
    make an intelligent decision
  • 00:08:55
    there yeah that it's it really is a
  • 00:08:57
    great question so
  • 00:09:00
    so I let me just try and bullet point
  • 00:09:03
    this so to speak okay so bullet point
  • 00:09:07
    number one don't reinvent the wheel um
  • 00:09:11
    unless you have a very specific reason
  • 00:09:13
    to you know reinvent a database why
  • 00:09:17
    would you I mean there's probably
  • 00:09:20
    there's probably someone out there who
  • 00:09:22
    needs their own database that isn't one
  • 00:09:25
    of the ones that we know and there's
  • 00:09:27
    good reason to do it that that's like
  • 00:09:29
    sort of really on the edge right it's
  • 00:09:31
    and and I would expect that the same
  • 00:09:34
    sort of rule applies to most software if
  • 00:09:36
    we're talking about business software so
  • 00:09:38
    there's you you really so this is at the
  • 00:09:41
    end of the day this is about cost of
  • 00:09:43
    ownership and Technical depth that
  • 00:09:45
    you're taking on right so if if you're
  • 00:09:48
    really starting from scratch I would say
  • 00:09:51
    you know um the pros and cons would
  • 00:09:54
    be obviously the flexibility and the
  • 00:09:57
    total control that starting from scratch
  • 00:09:58
    gives you
  • 00:10:00
    right versus the technical depth that
  • 00:10:04
    you're going to end up owning and and
  • 00:10:07
    maintaining and then all of this has to
  • 00:10:09
    kind of get get sort of factored into
  • 00:10:11
    some Financial projection in terms of
  • 00:10:13
    cost of ownership and you know what are
  • 00:10:15
    the true costs of this over time I would
  • 00:10:18
    so that's that's one the other
  • 00:10:22
    is systems don't live in isolation so if
  • 00:10:26
    you're thinking about a particular
  • 00:10:27
    system today don't just think of it as
  • 00:10:30
    as the one system but also think you
  • 00:10:32
    know what what other Integrations what
  • 00:10:35
    other systems do I need to build later
  • 00:10:37
    or interact with later and so all of
  • 00:10:39
    those things need to be taken into
  • 00:10:41
    consideration um I think for Enterprise
  • 00:10:44
    use cases
  • 00:10:47
    particularly starting from scratch
  • 00:10:50
    probably don't doesn't make sense for
  • 00:10:52
    most I would say it uh use cases that
  • 00:10:56
    are there yeah makes a lot of
  • 00:11:00
    um
  • 00:11:02
    so you you've you've built a whole
  • 00:11:04
    company out of um you know basically
  • 00:11:08
    doing the dev SEC Ops process around
  • 00:11:10
    service now
  • 00:11:13
    um and you you bring up interesting
  • 00:11:15
    points as to the you know hey they own
  • 00:11:17
    some of the code you're going to own
  • 00:11:18
    some of your own code and everything
  • 00:11:19
    like that um I guess one one kind of
  • 00:11:23
    question is what say to people that are
  • 00:11:28
    kind of like oh well um and maybe this
  • 00:11:31
    is a controversial question uh hey you
  • 00:11:34
    know we already spent all this money on
  • 00:11:35
    service now we're just going to kind of
  • 00:11:37
    go on our own and you know not worry so
  • 00:11:40
    much about the the dev seots Parts what
  • 00:11:43
    what considerations are they not making
  • 00:11:44
    and what challenges might they
  • 00:11:47
    face so let me just make sure I
  • 00:11:49
    understand the question you're asking me
  • 00:11:50
    if someone would tell me hey we're
  • 00:11:51
    thinking about getting off service now
  • 00:11:54
    no there there's someone who is like Hey
  • 00:11:56
    we're about to adopt service now um
  • 00:11:59
    we're not we're not going to utilize any
  • 00:12:00
    other tool sets or anything like that
  • 00:12:02
    we're just going to use us only what
  • 00:12:03
    service now gives us to to get the job
  • 00:12:06
    done oh well if you can do that that's
  • 00:12:08
    fine if you can do it just you know if
  • 00:12:10
    if if service now covers 100% of your
  • 00:12:13
    requirements and you're just happy
  • 00:12:14
    without of the Box you know I I wouldn't
  • 00:12:17
    push you to change anything and and and
  • 00:12:20
    so absolutely um and Serv now does do
  • 00:12:23
    that for for for a majority by the way
  • 00:12:26
    that that is the I think the main reason
  • 00:12:28
    like you know if you ask yourself what's
  • 00:12:30
    the real value of of a platform as a
  • 00:12:34
    service like service now like Salesforce
  • 00:12:36
    is that they've actually
  • 00:12:39
    built 90% or more the common
  • 00:12:42
    denominators of you know business
  • 00:12:44
    requirements X managing a sales a CRM
  • 00:12:47
    system or an IT management system or a
  • 00:12:49
    ticketing system right so you know
  • 00:12:51
    they've got that covered writing I I
  • 00:12:54
    would argue writing a CRM system from
  • 00:12:56
    scratch today probably doesn't make any
  • 00:12:58
    sense there's you know you need to have
  • 00:13:00
    a really good reason to build a new one
  • 00:13:03
    uh same I would say is is right for
  • 00:13:04
    service now if you're going to build a
  • 00:13:05
    new IT service management system or you
  • 00:13:08
    you know you'd ask yourself the same
  • 00:13:10
    question
  • 00:13:11
    so so that's the reason most customers
  • 00:13:14
    buy them now you ask so how come they
  • 00:13:16
    still have to customize it and they
  • 00:13:18
    still have to build well custom and I
  • 00:13:20
    want to start with the customization
  • 00:13:21
    piece which is the little bit is that
  • 00:13:24
    while the common functionality is the
  • 00:13:27
    same I have to map servers and
  • 00:13:29
    applications and networks and users and
  • 00:13:31
    and whatever that's the same regardless
  • 00:13:33
    of if you're in an the IT department of
  • 00:13:36
    an airline versus a bank versus a you
  • 00:13:39
    know a big retail store that's the same
  • 00:13:44
    but the details are different so so
  • 00:13:46
    one's going to have you know airplane
  • 00:13:48
    parts and the other one's going to have
  • 00:13:50
    banking systems so there is no way
  • 00:13:54
    around customizing it basically making
  • 00:13:56
    it your own and so that's layer number
  • 00:13:59
    one usually that's not so you know
  • 00:14:02
    that's not so Dev heavy or it could be
  • 00:14:07
    however once that's put in place where
  • 00:14:10
    Dev kind of gets layed on is that these
  • 00:14:13
    organizations see the platform they see
  • 00:14:17
    the benefit of how quickly they can get
  • 00:14:19
    something running because I mean by the
  • 00:14:22
    way that I I should have said that
  • 00:14:24
    before you're going to get stuff done
  • 00:14:26
    way way faster if you're on a platform
  • 00:14:28
    that is covered you know 80% of what you
  • 00:14:31
    needed to do scaffolding wise right um
  • 00:14:36
    and and going back to my point of now
  • 00:14:38
    that you you have something that's
  • 00:14:40
    working that same data data model could
  • 00:14:42
    be used say for customer service so
  • 00:14:45
    instead of building my customer service
  • 00:14:47
    application from scratch I'm actually
  • 00:14:49
    going to build my customer service on
  • 00:14:51
    the service now platform as well because
  • 00:14:53
    I'm using the same data model and I have
  • 00:14:55
    access to the same databases now you've
  • 00:14:58
    got both your thinking oh wait a minute
  • 00:15:00
    you know I've got um I've got
  • 00:15:04
    some HR needs in the company
  • 00:15:07
    right same same thing I can start it
  • 00:15:10
    from scratch and then integrate it into
  • 00:15:12
    service now because that's got the
  • 00:15:14
    catalog of you know who has what flaptop
  • 00:15:17
    and what what have you or I could build
  • 00:15:19
    the HR System on top or some of it or
  • 00:15:23
    some of the HR workflows on top of
  • 00:15:24
    service now right so it it just it just
  • 00:15:29
    makes a lot of sense for organizations
  • 00:15:32
    to do a lot of the a lot of these
  • 00:15:35
    Enterprise app apps on a single platform
  • 00:15:40
    single data
  • 00:15:41
    model yeah makes sense so uh another
  • 00:15:45
    another qu and you know you'll you'll
  • 00:15:47
    have some good insights and everything
  • 00:15:48
    here so uh obviously lots of people are
  • 00:15:50
    coming to you all and um frankly lots of
  • 00:15:53
    people come to us over at harness with
  • 00:15:56
    we are going to do more service or we're
  • 00:15:58
    doing lot of service now today we want
  • 00:16:00
    to do more service now um we're seeing
  • 00:16:02
    challenges in that devops process that
  • 00:16:05
    cicd process um you know frankly I'm
  • 00:16:08
    excited to have you on because again
  • 00:16:10
    I've only ever built service now tools
  • 00:16:12
    or anything in the service now UI and
  • 00:16:15
    done the uh you know the update sets and
  • 00:16:17
    everything like that and that's about as
  • 00:16:18
    far as I got so you know 101 what what
  • 00:16:22
    are the tools that are going to be
  • 00:16:24
    available to you know others in the
  • 00:16:28
    space uh obviously there's very
  • 00:16:30
    comprehensive out of the box things like
  • 00:16:32
    what you have at xtype but if you were
  • 00:16:34
    to go even lower and just say hey if if
  • 00:16:37
    you did have to do this um by yourself
  • 00:16:39
    like what's going to be available to you
  • 00:16:41
    are there interfacing tools are there
  • 00:16:43
    apis like how would I how would I eat
  • 00:16:46
    this whale so to speak yeah so if you're
  • 00:16:51
    embarking on you know development on the
  • 00:16:54
    service now platform and you're thinking
  • 00:16:56
    about
  • 00:16:57
    cicd um
  • 00:16:59
    well number one is service now has lots
  • 00:17:04
    lots of many different tools for
  • 00:17:06
    building
  • 00:17:07
    applications uh there's a workflow
  • 00:17:10
    Studio there's a a studio development
  • 00:17:13
    app which you can connect to get and
  • 00:17:16
    have all your code sort of stored in in
  • 00:17:19
    in a repository there's a there are new
  • 00:17:22
    there are new types of apps coming to
  • 00:17:24
    the service now platform there's even a
  • 00:17:26
    new programming language for fluent
  • 00:17:28
    which I'm I'm is is out yet or not that
  • 00:17:31
    will enable customers to build very
  • 00:17:33
    quickly and that and these are what's
  • 00:17:36
    called Pro code development on service
  • 00:17:39
    now then there is also the whole set of
  • 00:17:40
    low code tools that are there that you
  • 00:17:42
    know but whichever one you touch at the
  • 00:17:46
    end of the day you are making changes to
  • 00:17:49
    a particular service now instance when
  • 00:17:52
    you're developing regardless of whether
  • 00:17:54
    you're pushing code into gate or not and
  • 00:17:56
    then you're going to have to move your
  • 00:17:57
    changes or that new code from the
  • 00:18:00
    instance that you have used to develop
  • 00:18:03
    to the production instance and there may
  • 00:18:05
    be intermediate stages QA prepro and
  • 00:18:08
    what have you okay so how do you how do
  • 00:18:12
    you manage those those movements well
  • 00:18:16
    you can script them um and just you know
  • 00:18:20
    write your own scripts you can go out
  • 00:18:23
    and find um some cicd solutions that are
  • 00:18:27
    out there there's a
  • 00:18:29
    a really good guy in the service now
  • 00:18:31
    ecosystem called Sasha V gr who has uh
  • 00:18:35
    put out a whole very comprehensive set
  • 00:18:37
    of apis and white paper how you can
  • 00:18:40
    actually do push and pull uh between
  • 00:18:43
    instances and have Federated development
  • 00:18:45
    it's very interesting stuff you're going
  • 00:18:46
    to have to build it all yourself
  • 00:18:47
    basically right um and then what X Type
  • 00:18:51
    does if I you know if I uh
  • 00:18:54
    um if I may kind of just give some color
  • 00:18:57
    in what we do we we our aim is to is to
  • 00:19:01
    cover the wide spectrum so large
  • 00:19:04
    Enterprises will have development of All
  • 00:19:06
    Sorts happening on the platform and so
  • 00:19:09
    over time you've got multiple instances
  • 00:19:12
    that are basically in different states
  • 00:19:14
    all the time because development's
  • 00:19:15
    happening here it's happening there it's
  • 00:19:17
    happening here and happening on there
  • 00:19:19
    there's multiple of these instances but
  • 00:19:20
    at the end you're trying to flow all
  • 00:19:22
    that work into one production system
  • 00:19:24
    another complexity is that you're
  • 00:19:26
    developing different things that all
  • 00:19:28
    flow into one and that's where it kind
  • 00:19:30
    of gets hard to walk away from the need
  • 00:19:33
    to have a central control plane which
  • 00:19:35
    actually shows you differences and gives
  • 00:19:37
    you Automation and and tools to help you
  • 00:19:40
    manage those instances in the states
  • 00:19:44
    right exactly I I'm wondering if there's
  • 00:19:46
    uh lessons to be learned from the DB
  • 00:19:48
    admin Community you know we we have
  • 00:19:50
    a you know little little kind of sidebar
  • 00:19:53
    as well you know we're seeing a lot of
  • 00:19:54
    interest around database governance and
  • 00:19:57
    everything and service now I'm seeing a
  • 00:19:58
    lot of similarities between like what
  • 00:20:01
    we're seeing at a liquid base they've
  • 00:20:02
    got the concept of the update set liquid
  • 00:20:04
    base uses change sets and things like
  • 00:20:06
    that so uh maybe there's Lessons Learned
  • 00:20:08
    across that Community for hey you know
  • 00:20:10
    it's effectively a very regimented set
  • 00:20:13
    of instructions that are then done on
  • 00:20:15
    you know ident effectively logically
  • 00:20:18
    identical machines so uh as long as
  • 00:20:20
    those two things are true you can have a
  • 00:20:22
    multi-environment kind of deployment
  • 00:20:24
    strategy or synchronization strategy or
  • 00:20:27
    anything like that yeah you actually
  • 00:20:29
    well actually you touched you touched it
  • 00:20:31
    exactly it is we are we're talking about
  • 00:20:33
    we're talking consistency and
  • 00:20:34
    synchronization basically across
  • 00:20:36
    environments that are different but I
  • 00:20:38
    but but you but you you touched on
  • 00:20:40
    another very profound thing and I kind
  • 00:20:42
    of whizzed over it before but another
  • 00:20:45
    one of the big differences is that the
  • 00:20:47
    the database is given when you're on the
  • 00:20:51
    platform service now has selected the
  • 00:20:53
    database for you it's it's the service
  • 00:20:56
    now database of course that has a
  • 00:20:58
    relational database underneath it but
  • 00:20:59
    you have no access to the database
  • 00:21:01
    itself so so when you get a platform as
  • 00:21:05
    a service you're not just getting the
  • 00:21:08
    the hardware that's shared like an AWS
  • 00:21:10
    virtual machine or or physical machine
  • 00:21:13
    don't think those exist anymore but um
  • 00:21:15
    but you're actually getting the full
  • 00:21:17
    stack you're getting the database you're
  • 00:21:18
    getting the application layer you're
  • 00:21:20
    getting the apis it's all done for you
  • 00:21:23
    so your access is from the Top If you
  • 00:21:25
    will or the bottom through these things
  • 00:21:28
    and every change that you make every
  • 00:21:30
    customization or new feature that you
  • 00:21:32
    introduce will drag with it underlying
  • 00:21:37
    changes the database but you're not
  • 00:21:38
    going to be controlling those directly
  • 00:21:41
    because you're in in the platform so the
  • 00:21:44
    way that the platform manages changes
  • 00:21:46
    liquid base uh Salesforce has the same
  • 00:21:49
    thing Salesforce got I believe
  • 00:21:51
    Salesforce is also called a change set
  • 00:21:53
    um and service now has update sets
  • 00:21:56
    they're basically the same things the
  • 00:21:57
    records have changed
  • 00:21:59
    that are kept and then there's a there's
  • 00:22:02
    a an engine on the other instance that
  • 00:22:05
    knows how to read and Implement those
  • 00:22:06
    changes for you right so by the way that
  • 00:22:10
    was one of the big aha moments for me is
  • 00:22:12
    because when I thought of devops to me
  • 00:22:14
    devops always starts with Git and just
  • 00:22:18
    you know push and pull and merge and and
  • 00:22:21
    and that's how you know uh and that's
  • 00:22:24
    how federative development just all kind
  • 00:22:26
    of EVs and flows and merges and splits
  • 00:22:28
    and merges and splits until it you know
  • 00:22:30
    until some point in the at the end where
  • 00:22:32
    it just finally merges into this one
  • 00:22:34
    thing that works and on a platform you
  • 00:22:39
    don't have that because there's codebase
  • 00:22:41
    that's yours and codebase that's the
  • 00:22:43
    other so instead of trying to force that
  • 00:22:46
    model on the platform what we did is we
  • 00:22:48
    found a way to use the native facilities
  • 00:22:53
    those update sets as the core base to to
  • 00:22:58
    monit to them to show their exact status
  • 00:23:01
    across different instances to
  • 00:23:03
    synchronize them between instances to
  • 00:23:06
    allow developers to put policies on them
  • 00:23:09
    cicd type policies when this updates set
  • 00:23:12
    moves I want you to run that test before
  • 00:23:14
    this update it moves I need you to run
  • 00:23:16
    those scans right it's very cicd
  • 00:23:19
    resembling it's just built in the
  • 00:23:21
    platform same use case just build in the
  • 00:23:24
    platform versus trying to build it out
  • 00:23:26
    of the platform I think what one
  • 00:23:29
    question I have and you bring up a a we
  • 00:23:31
    kind of talk through a great point which
  • 00:23:33
    is like you know de devops is much
  • 00:23:36
    easier when the environments you
  • 00:23:38
    ultimately want to manage the changes
  • 00:23:40
    between are effectively nearly identical
  • 00:23:43
    right and you know uh my background is a
  • 00:23:47
    lot a lot of federal and the challenge
  • 00:23:48
    there is that you know what you do is
  • 00:23:51
    you build and test and deploy uh Dev
  • 00:23:55
    effectively deving QA versions of your
  • 00:23:57
    application and what we call the Lo side
  • 00:23:59
    um and then you know our customers
  • 00:24:01
    operate in What's called the high side
  • 00:24:03
    AKA like the secret environments the top
  • 00:24:05
    secret environments Etc um and then one
  • 00:24:08
    of the challenges we see there is like
  • 00:24:11
    hey you're on kubernetes the customers
  • 00:24:14
    on kubernetes you're on Amazon web
  • 00:24:16
    services the customer on Amazon web
  • 00:24:18
    services it should all be the same right
  • 00:24:20
    but inevitably there's something that is
  • 00:24:24
    mismatched that causes an issue and you
  • 00:24:28
    know triaging that are you guys seeing
  • 00:24:30
    the same thing in the service now SAS
  • 00:24:33
    environment or is that far less or tell
  • 00:24:35
    me a bit about that no it's it's it's
  • 00:24:38
    exactly that it is it is exactly
  • 00:24:42
    that we we're seeing it what you're
  • 00:24:44
    talking about is inconsistencies between
  • 00:24:47
    environments and those inconsistencies
  • 00:24:49
    inevitably built in an environment like
  • 00:24:52
    we're working in because you have
  • 00:24:55
    multiple teams right who are all
  • 00:24:59
    you know building on the same
  • 00:25:02
    environment and you could just I mean so
  • 00:25:05
    far we just talked about the code okay
  • 00:25:07
    how do we handle the code moving the
  • 00:25:08
    code testing the code synchronizing the
  • 00:25:10
    code across these multiple teams but
  • 00:25:13
    then there are stop there's also stuff
  • 00:25:15
    that that isn't code related uh it could
  • 00:25:17
    be data just different data and the D
  • 00:25:20
    and the data sometimes we kind of tend
  • 00:25:21
    to think data is irrelevant for code
  • 00:25:23
    well it's not because a lot of times the
  • 00:25:24
    data is you know like a becomes a
  • 00:25:27
    variable that that decide something and
  • 00:25:29
    now if you have a different data set the
  • 00:25:31
    application behaves differently uh
  • 00:25:33
    that's kind of rare but then there is
  • 00:25:35
    other things like a Plugin or a
  • 00:25:37
    third-party app that I use in my app
  • 00:25:40
    that you know has carries with it a
  • 00:25:42
    different version of kubernetes a
  • 00:25:44
    different version of a different version
  • 00:25:47
    of some API that's also used by another
  • 00:25:49
    team only those two teams don't know
  • 00:25:51
    they don't know it right you don't you
  • 00:25:53
    don't know it because you're working on
  • 00:25:55
    your own things so there's so and and
  • 00:25:57
    and it is that point that you have
  • 00:25:59
    multiple streams again different code
  • 00:26:01
    bases that are not connected not related
  • 00:26:04
    that are all trying to flow into the
  • 00:26:06
    same thing
  • 00:26:08
    across inconsistent environments and
  • 00:26:10
    then inevitably higher production let's
  • 00:26:13
    call it right is where they all have to
  • 00:26:16
    kind of align so where we start is with
  • 00:26:19
    just monitoring the differences
  • 00:26:21
    including production and and we just
  • 00:26:24
    start by showing the Deltas and that
  • 00:26:26
    also you know talk a little bit of Ai
  • 00:26:28
    and predictions Etc it helps us kind of
  • 00:26:31
    be proactively saying if you push this
  • 00:26:34
    code the chances of this working over
  • 00:26:37
    there or are either 100% nogo because we
  • 00:26:40
    can tell or we can you know say with
  • 00:26:43
    very high like they go this is not going
  • 00:26:45
    to work because of ab andc and we flag
  • 00:26:46
    it right yeah you bring up a good point
  • 00:26:49
    which is like even if the underlying
  • 00:26:50
    platforms and environments are exactly
  • 00:26:52
    the same the data is going to be
  • 00:26:54
    different high side versus low side and
  • 00:26:56
    if high
  • 00:26:57
    side
  • 00:26:58
    configurations Stu like that they'll
  • 00:27:00
    have a different ldap configuration
  • 00:27:02
    they'll have a different arback you know
  • 00:27:05
    SSO permissions and roles you know that
  • 00:27:07
    role is not there you all kinds of
  • 00:27:09
    little things that aren't code so we
  • 00:27:12
    have to take care of the code but we
  • 00:27:13
    also have to take care of our Rance so
  • 00:27:16
    how so how do this is going to be a
  • 00:27:19
    challenge consistent across every major
  • 00:27:20
    Enterprise every major massive Federal
  • 00:27:23
    organization how from your background
  • 00:27:26
    and everything in devops how should all
  • 00:27:28
    of these customers and everybody think
  • 00:27:30
    about solving that problem solving about
  • 00:27:34
    solving the like we want consistent
  • 00:27:36
    deployment near aor free across all
  • 00:27:40
    these different
  • 00:27:41
    environments yeah so we we did think
  • 00:27:43
    about it and we and we went ahead and we
  • 00:27:45
    implemented it and it it boils down in
  • 00:27:48
    the end to three main capabilities first
  • 00:27:50
    of which is visibility or monitoring
  • 00:27:53
    which you know usually isn't thought of
  • 00:27:55
    as as as a starting point for demop it's
  • 00:27:57
    like we're going to monitor performance
  • 00:27:59
    we're going to monitor user activity
  • 00:28:01
    transactions Etc it's mostly associated
  • 00:28:03
    with uh uh production use cases the
  • 00:28:06
    visibility I'm talking about is
  • 00:28:08
    visibility across environments it's
  • 00:28:10
    basically something that replaces your
  • 00:28:12
    git repo as the single source of Truth
  • 00:28:16
    and it's and it it is a a a monitoring
  • 00:28:18
    capability that doesn't just look at
  • 00:28:21
    code it looks at code is number one
  • 00:28:23
    across those different pillars but it
  • 00:28:25
    also looks at the stuff we talked about
  • 00:28:26
    looks at data differences configuration
  • 00:28:28
    differences plugins and other things
  • 00:28:31
    that may be installed in one environment
  • 00:28:32
    or not the other maybe are in different
  • 00:28:34
    versions Etc so that's visibility that's
  • 00:28:37
    number one then the second part is
  • 00:28:41
    around automation of anything that you
  • 00:28:44
    can and this is very devops uh style
  • 00:28:47
    right um make sure that the tests run in
  • 00:28:50
    in concurrently to any code pushes make
  • 00:28:54
    sure that your you know your unit test
  • 00:28:56
    your your your your code scans all of
  • 00:28:58
    that like the the cicd functionality is
  • 00:29:01
    all automated automate the release into
  • 00:29:04
    production itself whether it's creating
  • 00:29:05
    packages and automating those
  • 00:29:07
    integrating them into some sort of
  • 00:29:09
    schedules but then use that automation
  • 00:29:12
    to also create and that's the third
  • 00:29:14
    pillar of governance usually we say
  • 00:29:16
    visibility governance and automation you
  • 00:29:18
    can you know you can it always starts
  • 00:29:20
    with with visibility and by governance I
  • 00:29:23
    mean
  • 00:29:24
    is governance is is where
  • 00:29:29
    you take these cicd like
  • 00:29:33
    automations but you hook them up to
  • 00:29:35
    triggers and and to to avoid consistency
  • 00:29:39
    increase compliance and let me give you
  • 00:29:41
    an example so if A Change Is flown from
  • 00:29:44
    one instance into a QA instance and it's
  • 00:29:46
    now in QA and ready now I know that the
  • 00:29:49
    QA instance and this Dev instance are
  • 00:29:51
    inconsistent with my
  • 00:29:53
    third so I can have a governance rule
  • 00:29:56
    that says anything that makes it to QA
  • 00:29:58
    and is successfully committed needs to
  • 00:30:00
    flow down needs to go back to a
  • 00:30:03
    different development environment so
  • 00:30:04
    they are so they remain consistent you
  • 00:30:06
    got to be smart about it you don't
  • 00:30:08
    necessarily always want to push
  • 00:30:09
    everything but you have to think about
  • 00:30:11
    that what what do I need to synchronize
  • 00:30:14
    you should do the same for data right
  • 00:30:17
    you should do the same for for
  • 00:30:19
    configurations you can do you should do
  • 00:30:21
    the same for uh uh for for plugins or
  • 00:30:24
    different apps Etc um and that's
  • 00:30:27
    basically what what we said his ability
  • 00:30:29
    governance and Automation and how you do
  • 00:30:31
    it whether you implement it with an open
  • 00:30:33
    source tool you build it yourself you
  • 00:30:35
    buy a product that that's on you yeah
  • 00:30:38
    very nice um so you've been you've been
  • 00:30:41
    around the block a few times in the
  • 00:30:42
    devops community this will be kind of a
  • 00:30:45
    two-part set of questions but I'll I'll
  • 00:30:46
    go one by one I mean first off is um
  • 00:30:51
    what what are the devops challenges that
  • 00:30:53
    have remained consistent and haven't
  • 00:30:54
    been solved since you've since you
  • 00:30:56
    started working on these you know over
  • 00:30:58
    two decades
  • 00:31:03
    ago all right
  • 00:31:07
    um devops is hard that has not changed
  • 00:31:13
    uh I was there when the Mainframe was
  • 00:31:16
    around but when I got into the industry
  • 00:31:18
    my elders were constantly telling me
  • 00:31:21
    about devops on the main frame and I
  • 00:31:22
    remember it very well and this was one
  • 00:31:24
    big computer right and they still had
  • 00:31:26
    challenges and then we went kind of
  • 00:31:30
    distributed and we went into we tried to
  • 00:31:32
    break up the the the world into smaller
  • 00:31:35
    components that interact with with each
  • 00:31:37
    other believing that that the that this
  • 00:31:40
    this comp you know
  • 00:31:43
    componentization
  • 00:31:45
    comp compiz
  • 00:31:48
    sorry yeah cut that out the
  • 00:31:50
    componentization would help and it just
  • 00:31:52
    created a myriad of objects and and so
  • 00:31:54
    that created a whole other mess we had
  • 00:31:57
    right um we had waterfall that was meant
  • 00:32:00
    to kind of ensure that the projects are
  • 00:32:02
    up to standard but then then you know
  • 00:32:05
    you only do one jump into production so
  • 00:32:08
    and and that didn't work so we went
  • 00:32:10
    agile and we broke up the process and we
  • 00:32:12
    decided that there is no more release
  • 00:32:14
    day it's it's just ongoing at the end of
  • 00:32:17
    the day the challenge of building
  • 00:32:20
    software and operating it is big and as
  • 00:32:25
    complexity grows that remains if I want
  • 00:32:27
    to be more more specific I think the the
  • 00:32:30
    premise that you have a production
  • 00:32:32
    version that is live serving customers
  • 00:32:36
    and other versions that are being built
  • 00:32:38
    the whole time that challenge is never
  • 00:32:41
    going going away regardless of whether
  • 00:32:43
    you choose a different architecture you
  • 00:32:45
    choose a different a different uh
  • 00:32:47
    development methodology you're going to
  • 00:32:49
    work with scrum or you're going to use
  • 00:32:51
    you know agile ports and and and
  • 00:32:53
    waterfalls even Etc the end of the day
  • 00:32:56
    you're you're dealing with
  • 00:32:58
    change in complex systems and so
  • 00:33:01
    managing that is just it it's a
  • 00:33:03
    challenge so you're always going to have
  • 00:33:05
    that challenges it hasn't changed put it
  • 00:33:07
    that way production still
  • 00:33:09
    production I mean have you met those
  • 00:33:11
    that have you met individuals that do
  • 00:33:13
    believe that like hey we don't need a
  • 00:33:16
    Dev QA in prod we we just do everything
  • 00:33:18
    in prod you know changes go directly
  • 00:33:21
    there I mean I have met those people but
  • 00:33:23
    um I don't know I'm curious your
  • 00:33:25
    experiences it's funny I I I actually
  • 00:33:28
    run into quite a lot of them and but
  • 00:33:31
    this is very specific to to I don't know
  • 00:33:34
    if it's just specific to service now but
  • 00:33:36
    spe I I know it from service now so when
  • 00:33:38
    you start out early on a platform like
  • 00:33:40
    this things just look so easy it is so
  • 00:33:43
    easy to just create a new app like I
  • 00:33:45
    will write you a new app within 10
  • 00:33:47
    minutes It'll be up and running and you
  • 00:33:49
    know what it it's going to handle a
  • 00:33:51
    thousand people you know uh registering
  • 00:33:54
    on it 10,000 people so you're Superman
  • 00:33:56
    right and you just do it on
  • 00:33:58
    production and then as things scale up a
  • 00:34:02
    little bit people are just falling into
  • 00:34:04
    this thing so you know what one of the
  • 00:34:05
    one of the issues is that developers are
  • 00:34:07
    always
  • 00:34:08
    admins right and so they're used to it
  • 00:34:12
    so if something is wrong I'm just going
  • 00:34:13
    to go into production and fix my thing
  • 00:34:15
    where this comes back to bite you is
  • 00:34:18
    obviously later as you grow right and
  • 00:34:21
    now you have a much more complex
  • 00:34:22
    environment those habits call them that
  • 00:34:26
    way become very detrimental so you're
  • 00:34:30
    seeing so I see a whole spectrum of
  • 00:34:33
    organizations I see organizations you
  • 00:34:34
    know large especially large Banks and
  • 00:34:36
    institutions who would never allow it
  • 00:34:38
    because they're they're they are they so
  • 00:34:41
    from day one they have very stringent
  • 00:34:43
    processes Etc which creates another
  • 00:34:46
    problem they move very very slow unless
  • 00:34:47
    they automate their processes Etc and
  • 00:34:49
    then on the flip side I see
  • 00:34:51
    organizations that kind of grew and then
  • 00:34:53
    as they got to scale they still have
  • 00:34:55
    those old habits of people just going
  • 00:34:56
    into production and changing things and
  • 00:34:58
    doing things and messing everything
  • 00:35:00
    up yeah no I mean uh I'm going to I'm
  • 00:35:04
    going to clip this out and be like hey
  • 00:35:06
    for all all you all you customers out
  • 00:35:07
    there that think like hey prod is Dev um
  • 00:35:10
    you know maybe let's uh maybe it's easy
  • 00:35:12
    today but it's not going to be so easy
  • 00:35:14
    tomorrow so let's keep that in
  • 00:35:16
    mind I yeah so so I mentioned this was
  • 00:35:20
    kind of a two-parter question so and it
  • 00:35:22
    was a fascinating answer for what it's
  • 00:35:24
    worth the idea of like um and frankly
  • 00:35:26
    not one I expected the idea of like the
  • 00:35:29
    dev QA prod kind of model or even just a
  • 00:35:32
    Dev prod model has been a - tested
  • 00:35:35
    battle hardened model that's worked for
  • 00:35:38
    devops for for basically since the
  • 00:35:40
    beginning um the other question though
  • 00:35:43
    is uh we're in an age where there's AI
  • 00:35:48
    there's you know lots of new
  • 00:35:49
    methodologies for automation you know
  • 00:35:52
    what's what's the future of De Su Ops or
  • 00:35:54
    what are the pieces of the future of De
  • 00:35:56
    Dev Su Ops that maybe you're the most
  • 00:35:58
    excited about or or that you have
  • 00:36:00
    insight
  • 00:36:01
    into yeah
  • 00:36:04
    so I am a a techno Optimist and and a
  • 00:36:07
    techno lover and so obviously like
  • 00:36:09
    everybody else you know I I I spend a
  • 00:36:12
    lot of time trying to learn we learning
  • 00:36:14
    about you know llms and how they work
  • 00:36:17
    and you know what models look like and
  • 00:36:19
    you think about the use cases that are
  • 00:36:21
    there we we're all familiar with chat
  • 00:36:22
    GPT it's kind of taking us on and so
  • 00:36:24
    we're just focused on on just text
  • 00:36:28
    uh Us in the dev Community we all we're
  • 00:36:30
    all familiar with you know with co-pilot
  • 00:36:32
    and others and see how um you know
  • 00:36:34
    coding is getting better on these
  • 00:36:37
    platforms but to me what's exciting is
  • 00:36:40
    the is the the fundamental notion that
  • 00:36:42
    these systems predict stuff based off
  • 00:36:44
    based off of you know um data that
  • 00:36:48
    they've seen and so one of the most
  • 00:36:51
    exciting things for me is to start
  • 00:36:53
    seeing the software life cycle be
  • 00:36:55
    predicted across everything so we can
  • 00:36:57
    start
  • 00:36:59
    from relative simple and I say it with
  • 00:37:01
    all the respect because I haven't seen a
  • 00:37:03
    good one working in simple use cases
  • 00:37:05
    that a lot of people can think of hey
  • 00:37:07
    I'm just going to monitor a codebase
  • 00:37:10
    right and the codebase is my learning
  • 00:37:12
    thing and then I'm going to have this AI
  • 00:37:14
    that just predicts the test so you don't
  • 00:37:16
    have to test you don't have to test you
  • 00:37:18
    know you can get the best coverage with
  • 00:37:20
    the minimum amount of test or it will
  • 00:37:22
    tell it will automatically decide to
  • 00:37:24
    only run this test for this release it
  • 00:37:26
    will automatically decide I'm only I'm
  • 00:37:27
    only I'm only going to need this ABC and
  • 00:37:29
    the rest is unnecessary and this will
  • 00:37:31
    work with you know with a very high
  • 00:37:33
    conviction rate or something like that
  • 00:37:34
    so you're going to see a lot of decision
  • 00:37:37
    making Automation in in the process and
  • 00:37:40
    I can I can see that sort of I believe
  • 00:37:44
    starts with the the use case that I just
  • 00:37:46
    talked about but you can expand that to
  • 00:37:48
    to everything so predicting what tests
  • 00:37:51
    are needed maybe writing them ahead of
  • 00:37:52
    time and then executing them and then
  • 00:37:55
    you can go wild with it say Hey you know
  • 00:37:58
    at the end of the day if we if we become
  • 00:38:00
    really autonomous then I'd like the
  • 00:38:02
    system to predict the data and the
  • 00:38:05
    metrics that the managers want and spit
  • 00:38:08
    that out and then if we go into sci-fi
  • 00:38:11
    then you know some AI can decide hey in
  • 00:38:13
    order for us to beat the competition we
  • 00:38:15
    need to develop this system and there we
  • 00:38:18
    go I've dream it I've created it right
  • 00:38:22
    like something constantly monitoring all
  • 00:38:24
    the news all the oh you know your
  • 00:38:26
    competitor just came out with this
  • 00:38:27
    feature uh let's prioritize they came
  • 00:38:29
    out with these features we don't think
  • 00:38:31
    there's going to be a lot of demand it
  • 00:38:33
    it came out with this one though this
  • 00:38:34
    one we think is killer based on Reddit
  • 00:38:36
    forums based on you know the user
  • 00:38:38
    Community Etc Bas based on our code base
  • 00:38:41
    based on our code base we have we have
  • 00:38:42
    an advantage we have an advantage in our
  • 00:38:44
    code base because we've got this thing
  • 00:38:45
    and it could be repurposed to do that
  • 00:38:47
    that will kill this crazy stuff like
  • 00:38:50
    that I think we're a long ways away I
  • 00:38:51
    think it's going to get it's going to
  • 00:38:52
    start with a lot more
  • 00:38:55
    nuanced you you bring up a so I I've
  • 00:38:59
    I've interacted a lot with customers and
  • 00:39:01
    I think uh it's very rare to meet
  • 00:39:03
    developers that love tdd test driven
  • 00:39:06
    development I think a lot of them just
  • 00:39:08
    want to write the code they don't want
  • 00:39:09
    to do so much with the test so I mean it
  • 00:39:11
    sounds like craigman from wrong you're
  • 00:39:13
    seeing a lot of potential in maybe just
  • 00:39:15
    recognizing that hey and and there's
  • 00:39:18
    almost like a governance piece to it of
  • 00:39:19
    like Developers for the most part don't
  • 00:39:21
    like writing tests so maybe instead of
  • 00:39:24
    like trying to convince them it's
  • 00:39:26
    important or anything like that just
  • 00:39:27
    recognize that's something they're never
  • 00:39:28
    going to like it's the it's the diet of
  • 00:39:31
    you know the the co coding industry and
  • 00:39:34
    let's just recognize that and use
  • 00:39:36
    automations Ai and everything to get
  • 00:39:38
    them where they need to go there and let
  • 00:39:39
    them focus on the value ad 100% 100%
  • 00:39:43
    yeah that's I that's a that's a good way
  • 00:39:46
    to think about it but got the governance
  • 00:39:48
    to make sure it happens but you've got
  • 00:39:49
    the automations and everything to you
  • 00:39:51
    know make it a a less stressful or less
  • 00:39:55
    impactful part of the day yeah and and
  • 00:39:56
    let the Vel opers do the stuff that they
  • 00:39:58
    love let them you know write the code
  • 00:40:00
    see it work think it design it you know
  • 00:40:03
    implement
  • 00:40:04
    it push it see that it works and then
  • 00:40:07
    forget about yeah I I want to double
  • 00:40:10
    click on governance because you know
  • 00:40:12
    that selfishly that's a huge part of
  • 00:40:14
    what we do at harness and that's a huge
  • 00:40:16
    part of what you guys are doing as well
  • 00:40:18
    um I think uh I'll I'll set this
  • 00:40:22
    question up a little bit before I
  • 00:40:24
    started working here when I heard
  • 00:40:26
    governance I thought paperwork I thought
  • 00:40:30
    you know oh it's a it's a review board
  • 00:40:32
    it's a it's a cab or something like that
  • 00:40:35
    um and I think that there's kind of
  • 00:40:39
    maybe a cultural change as well as you
  • 00:40:41
    know other kind of tech changes and
  • 00:40:43
    everything that are changing the
  • 00:40:44
    narrative there do you have any thoughts
  • 00:40:46
    on that or just curious your take yeah
  • 00:40:48
    I'll give you an anonymous example a
  • 00:40:50
    story okay so we were working with this
  • 00:40:54
    very very large bank and got introduced
  • 00:40:58
    to their service now team they have a
  • 00:41:01
    huge team over 300 developers just on
  • 00:41:03
    service now you know many many many
  • 00:41:07
    instances and a lot of governance this
  • 00:41:09
    is a bank regulated industry every
  • 00:41:11
    change you know there there's process
  • 00:41:13
    around it and they had governance
  • 00:41:15
    exactly like you described it right
  • 00:41:18
    paperwork tons of it they had paperwork
  • 00:41:21
    describing the paperwork right right and
  • 00:41:24
    then some more paperwork and then
  • 00:41:26
    paperwork that you got got to do it and
  • 00:41:28
    and that was the governance model and I
  • 00:41:32
    remember my dear friend hak who's you
  • 00:41:34
    know our chief strategy officer the time
  • 00:41:37
    we were looking at it when we got back
  • 00:41:40
    we were heading into a PC this was one
  • 00:41:42
    of our earliest customers and we you
  • 00:41:45
    know and what we and we got from them a
  • 00:41:50
    a power PowerPoint
  • 00:41:52
    presentation that was 82 slides long
  • 00:41:55
    that showed their process and end to end
  • 00:41:58
    and they basically said okay this is our
  • 00:42:00
    process how are you going to build this
  • 00:42:02
    with
  • 00:42:03
    xti because we're GNA try to do that and
  • 00:42:07
    I remember Huck sending me an email
  • 00:42:10
    going this is like we don't we
  • 00:42:13
    don't need any of this like out of these
  • 00:42:16
    80 slides there's probably only about
  • 00:42:18
    three that we need the rest we're we can
  • 00:42:20
    just strip away this is
  • 00:42:22
    all redundant unnecessary and I remember
  • 00:42:26
    saying to him if if we go back and tell
  • 00:42:29
    them that they're not going to talk to
  • 00:42:31
    us they're just they're just going to
  • 00:42:33
    ignore us so we have no choice my friend
  • 00:42:36
    we're going to go we're going to start
  • 00:42:38
    this way we're going to go into their
  • 00:42:40
    document we're going to show them
  • 00:42:41
    exactly where we fit into their existing
  • 00:42:44
    process right now and it was painstaking
  • 00:42:47
    and hard and honestly very frustrating
  • 00:42:51
    to do that but we did and we sent it
  • 00:42:55
    back and then we got in you know roll
  • 00:42:58
    forward a year and a half they're only
  • 00:43:01
    left with two
  • 00:43:03
    slides the rest just went
  • 00:43:05
    away so the so so the Theo the moral of
  • 00:43:09
    the story is
  • 00:43:10
    that with proper
  • 00:43:13
    tools devops can become governance and
  • 00:43:16
    should become governance um that goes
  • 00:43:19
    beyond just developers but for that you
  • 00:43:21
    have to enlist you know the people that
  • 00:43:23
    are sitting on top that are being liable
  • 00:43:25
    accountable for a lot of the stuff and
  • 00:43:27
    sometimes for them we have to work for
  • 00:43:30
    them too not just for again this goes
  • 00:43:32
    back to if we just love code that's fine
  • 00:43:36
    um but sometimes a lot of that other
  • 00:43:40
    stuff is not really for us it's to
  • 00:43:42
    satisfy other people with other
  • 00:43:44
    needs exactly no I mean that's that's an
  • 00:43:48
    incredible story and I I think the
  • 00:43:50
    lesson that maybe be learned out of that
  • 00:43:51
    is like you know if you have something
  • 00:43:54
    that is going to help customers in terms
  • 00:43:57
    of like making an organizational change
  • 00:43:59
    um or maybe like a culture change or
  • 00:44:01
    anything like that it could be important
  • 00:44:02
    to meet them where they are first and
  • 00:44:05
    then have the conviction that the change
  • 00:44:07
    will come but you know we have to we got
  • 00:44:10
    to meet them where they are to for what
  • 00:44:11
    it's worth so yeah I mean look we're
  • 00:44:13
    we're techies and and many times we
  • 00:44:16
    understand technology better than than
  • 00:44:18
    our customers that's why we do what we
  • 00:44:20
    do right that's where we are but you
  • 00:44:23
    have I think we have we better off
  • 00:44:25
    realizing that you can't normally just
  • 00:44:29
    throw some Noel idea on someone even if
  • 00:44:32
    you're very convinced and you know it
  • 00:44:33
    you go oh how can you be so part of my
  • 00:44:35
    friend I'm going be so stupid look let
  • 00:44:37
    me show you this that's not a very
  • 00:44:40
    effective way because the reality is
  • 00:44:42
    maybe they don't see what you see maybe
  • 00:44:44
    they have been burnt by other stuff
  • 00:44:46
    before so you know you you have
  • 00:44:48
    to sometimes you you have
  • 00:44:51
    to if you want to AFF change well
  • 00:44:53
    sometimes you have to start meet them
  • 00:44:55
    where they are and then do it over time
  • 00:44:57
    so we're we're up on time but I want to
  • 00:44:59
    hit you with like you know three or four
  • 00:45:01
    rapid fires just to like um you know get
  • 00:45:04
    get them out there and we can there yeah
  • 00:45:07
    so rapid fire round let let's go uh
  • 00:45:11
    you're Tech you do a lot what is the
  • 00:45:13
    most underrated tool in the devop space
  • 00:45:15
    right
  • 00:45:18
    now what is the most underrated tool in
  • 00:45:20
    the devop space right now
  • 00:45:24
    yeah Version Control version control not
  • 00:45:27
    love it what's a what's a buzz word that
  • 00:45:29
    you wish would go away
  • 00:45:34
    forever agile yeah we could we could
  • 00:45:36
    have a whole podcast just about that um
  • 00:45:39
    and then last one what is your go-to
  • 00:45:41
    language of choice for getting an
  • 00:45:43
    automation
  • 00:45:48
    done I hate to say it um but it's got to
  • 00:45:51
    be
  • 00:45:53
    JavaScript JavaScript I that's I think
  • 00:45:55
    that's the first I've heard it from
  • 00:45:57
    friends and everybody like that but my
  • 00:45:59
    favorite my favorite programming
  • 00:46:01
    language the language I I really love so
  • 00:46:04
    I grew up as C programmer so that's my
  • 00:46:07
    core I I did C for a long time and my
  • 00:46:10
    favorite language right now is go um and
  • 00:46:13
    I love go and I wish but when I want to
  • 00:46:16
    get stuff done quick it's JavaScript
  • 00:46:19
    JavaScript and you just throw that in
  • 00:46:20
    like a nodejs app and then that's a
  • 00:46:23
    command line tool nice I need to pick
  • 00:46:25
    more of that stuff up well it's h it's
  • 00:46:27
    at the top of the hour and I want to
  • 00:46:30
    thank you for hopping on ship talk and
  • 00:46:32
    it's been a pleasure any any kind of
  • 00:46:33
    final comments or or thoughts to the
  • 00:46:35
    audience no thank you very much if you
  • 00:46:38
    know if you're interested or if you have
  • 00:46:41
    questions about service now or platforms
  • 00:46:42
    or just think about that this Niche that
  • 00:46:45
    we're in just hit me up I'm always happy
  • 00:46:46
    to help and uh really appreciate your
  • 00:46:48
    time thank you Nick it's great right
  • 00:46:51
    fantastic enjoy your day Ron take care
  • 00:46:56
    [Music]
Etiquetas
  • DevOps
  • ServiceNow
  • Automation
  • Governance
  • Visibility
  • Multi-instance Management
  • AI
  • Software Delivery
  • Integration
  • Version Control