KISS - Keep It Simple Stupid | Software Architecture Conference

Name: KISS - Keep It Simple Stupid | Software Architecture Conference | Open Video
Uploaded: 2025-08-06T12:06:12+00:00
Duration: 35 min 59 s

0:03
um so my talk is keep it simple stupid
0:06
um so my talk is keep it simple stupid
0:06
um so my talk is keep it simple stupid uh this
0:07
uh this
0:07
uh this is this is really meta when it comes to
0:11
is this is really meta when it comes to
0:11
is this is really meta when it comes to development because this is really the
0:12
development because this is really the
0:12
development because this is really the process around um how you develop how
0:16
process around um how you develop how
0:16
process around um how you develop how you start a project from ground uh from
0:19
you start a project from ground uh from
0:19
you start a project from ground uh from point zero and get it to the end um and
0:22
point zero and get it to the end um and
0:22
point zero and get it to the end um and it's really about designing code for
0:24
it's really about designing code for
0:24
it's really about designing code for maintainability and collaboration um
0:26
maintainability and collaboration um
0:26
maintainability and collaboration um very rarely are we the only developer
0:28
very rarely are we the only developer
0:28
very rarely are we the only developer who ever works on something thing and uh
0:32
who ever works on something thing and uh
0:32
who ever works on something thing and uh especially if you work for a large
0:33
especially if you work for a large
0:33
especially if you work for a large organization a lot of times the code
0:36
organization a lot of times the code
0:36
organization a lot of times the code you're working on was not necessarily
0:37
you're working on was not necessarily
0:37
you're working on was not necessarily written by you and will be maintained
0:39
written by you and will be maintained
0:39
written by you and will be maintained and written by people Beyond um so a all
0:42
and written by people Beyond um so a all
0:42
and written by people Beyond um so a all of these things you know go into all of
0:45
of these things you know go into all of
0:45
of these things you know go into all of that um so what I'm going to prove today
0:47
that um so what I'm going to prove today
0:47
that um so what I'm going to prove today is that you're paid to solve problems
0:49
is that you're paid to solve problems
0:49
is that you're paid to solve problems and not write code um I I feel like this
0:52
and not write code um I I feel like this
0:52
and not write code um I I feel like this is common wisdom if you've been around
0:55
is common wisdom if you've been around
0:55
is common wisdom if you've been around for a while um but it it really Bears
0:58
for a while um but it it really Bears
0:58
for a while um but it it really Bears repeating and I think there's some
1:00
repeating and I think there's some
1:00
repeating and I think there's some things that once we start getting coding
1:02
things that once we start getting coding
1:02
things that once we start getting coding we kind of Miss on
1:04
we kind of Miss on
1:04
we kind of Miss on that another one is for every 1 hour
1:06
that another one is for every 1 hour
1:06
that another one is for every 1 hour Sprint PL spent planning you save 10
1:09
Sprint PL spent planning you save 10
1:09
Sprint PL spent planning you save 10 hours of
1:10
hours of
1:10
hours of execution um now that sounds crazy um
1:13
execution um now that sounds crazy um
1:13
execution um now that sounds crazy um but as I'll talk about a little bit
1:15
but as I'll talk about a little bit
1:15
but as I'll talk about a little bit later and showing examples it you cannot
1:18
later and showing examples it you cannot
1:18
later and showing examples it you cannot over plan and you truly can cut
1:22
over plan and you truly can cut
1:22
over plan and you truly can cut development times almost in half by just
1:24
development times almost in half by just
1:24
development times almost in half by just spending more time planning planning
1:27
spending more time planning planning
1:27
spending more time planning planning planning um and the last one is less is
1:31
planning um and the last one is less is
1:31
planning um and the last one is less is more so a bit about myself um I've been
1:35
more so a bit about myself um I've been
1:35
more so a bit about myself um I've been in software development for over 25
1:37
in software development for over 25
1:37
in software development for over 25 years I have programmed in almost every
1:41
years I have programmed in almost every
1:41
years I have programmed in almost every language professionally I've touched a
1:43
language professionally I've touched a
1:43
language professionally I've touched a whole bunch of languages just for fun uh
1:46
whole bunch of languages just for fun uh
1:46
whole bunch of languages just for fun uh I've spent five years plus on the w3c
1:49
I've spent five years plus on the w3c
1:50
I've spent five years plus on the w3c digital publishing group so I have
1:51
digital publishing group so I have
1:51
digital publishing group so I have experience in
1:53
experience in
1:53
experience in standards um I've worked for startups
1:56
standards um I've worked for startups
1:56
standards um I've worked for startups midsize companies and large siiz
1:57
midsize companies and large siiz
1:57
midsize companies and large siiz companies and through the years have
1:59
companies and through the years have
1:59
companies and through the years have been part of and designed and developed
2:02
been part of and designed and developed
2:02
been part of and designed and developed over 10 applications from the idea phase
2:05
over 10 applications from the idea phase
2:05
over 10 applications from the idea phase to reality so not just coming in and
2:07
to reality so not just coming in and
2:07
to reality so not just coming in and maintaining projects I've been part of
2:10
maintaining projects I've been part of
2:10
maintaining projects I've been part of four major large scale
2:13
four major large scale
2:13
four major large scale rewrites um I through much of my career
2:17
rewrites um I through much of my career
2:17
rewrites um I through much of my career have been a project firefighter which
2:19
have been a project firefighter which
2:19
have been a project firefighter which basically is whenever a project is
2:22
basically is whenever a project is
2:22
basically is whenever a project is failing I get thrown into it to try and
2:24
failing I get thrown into it to try and
2:24
failing I get thrown into it to try and fix it so I have come into a lot of
2:28
fix it so I have come into a lot of
2:28
fix it so I have come into a lot of messes and and uh through that I've
2:32
messes and and uh through that I've
2:32
messes and and uh through that I've learned kind of how to clean up and how
2:34
learned kind of how to clean up and how
2:34
learned kind of how to clean up and how to prevent those messes so that's really
2:35
to prevent those messes so that's really
2:35
to prevent those messes so that's really what I'm trying to get out to here get
2:38
what I'm trying to get out to here get
2:38
what I'm trying to get out to here get out to you and I am a Hands-On
2:40
out to you and I am a Hands-On
2:40
out to you and I am a Hands-On development manager a coach and a
2:44
speaker um now a little bit more about
2:46
speaker um now a little bit more about
2:46
speaker um now a little bit more about me I am not a 10x
2:48
me I am not a 10x
2:48
me I am not a 10x programmer um my code is my code is
2:52
programmer um my code is my code is
2:52
programmer um my code is my code is simple my code is elegant and it works
2:54
simple my code is elegant and it works
2:55
simple my code is elegant and it works um but I I met a few people that we call
2:58
um but I I met a few people that we call
2:58
um but I I met a few people that we call those 10x programmers that's not me um
3:01
those 10x programmers that's not me um
3:01
those 10x programmers that's not me um I'm just kind of some regular average
3:03
I'm just kind of some regular average
3:03
I'm just kind of some regular average guy who's just spent a lot of time doing
3:04
guy who's just spent a lot of time doing
3:04
guy who's just spent a lot of time doing this sort of stuff um I've over over
3:07
this sort of stuff um I've over over
3:07
this sort of stuff um I've over over engineered at least 50% of my projects
3:10
engineered at least 50% of my projects
3:10
engineered at least 50% of my projects so I've made enough mistakes that I can
3:11
so I've made enough mistakes that I can
3:11
so I've made enough mistakes that I can now talk about them and hopefully help
3:13
now talk about them and hopefully help
3:13
now talk about them and hopefully help other people prevent them I've
3:15
other people prevent them I've
3:15
other people prevent them I've accidentally dropped a table in prod so
3:18
accidentally dropped a table in prod so
3:18
accidentally dropped a table in prod so um if you've done that then you are in
3:20
um if you've done that then you are in
3:20
um if you've done that then you are in good company um I've written poor commit
3:22
good company um I've written poor commit
3:22
good company um I've written poor commit messages I have edited code directly in
3:25
messages I have edited code directly in
3:25
messages I have edited code directly in prod and I've spent hours trying to
3:28
prod and I've spent hours trying to
3:28
prod and I've spent hours trying to optimize something that wasn't even a
3:29
optimize something that wasn't even a
3:29
optimize something that wasn't even a performance issue so um you know I've
3:33
performance issue so um you know I've
3:33
performance issue so um you know I've I've seen the highs and I've seen the
3:36
I've seen the highs and I've seen the
3:36
I've seen the highs and I've seen the lows so this should be very familiar to
3:39
lows so this should be very familiar to
3:39
lows so this should be very familiar to people this is kind of that that
3:41
people this is kind of that that
3:41
people this is kind of that that one-year project you can squeeze it if
3:43
one-year project you can squeeze it if
3:44
one-year project you can squeeze it if you want to make it a six-year but when
3:45
you want to make it a six-year but when
3:46
you want to make it a six-year but when you're working for a mid to large size
3:48
you're working for a mid to large size
3:48
you're working for a mid to large size company you'll see this um and usually
3:51
company you'll see this um and usually
3:51
company you'll see this um and usually it starts with the business giving you a
3:54
it starts with the business giving you a
3:54
it starts with the business giving you a set of requirements you know we need an
3:55
set of requirements you know we need an
3:55
set of requirements you know we need an application that does XYZ ABC whatever
3:59
application that does XYZ ABC whatever
3:59
application that does XYZ ABC whatever and they hand it to the development team
4:03
and they hand it to the development team
4:03
and they hand it to the development team and at that point the development team
4:04
and at that point the development team
4:04
and at that point the development team has about one month for
4:07
has about one month for
4:07
has about one month for planning and then they have to start
4:09
planning and then they have to start
4:09
planning and then they have to start showing something so it's Off to the
4:12
showing something so it's Off to the
4:12
showing something so it's Off to the Races you develop for 10 months and then
4:14
Races you develop for 10 months and then
4:14
Races you develop for 10 months and then you've got one month for QA and user
4:16
you've got one month for QA and user
4:16
you've got one month for QA and user acceptance testing and then you
4:19
acceptance testing and then you
4:19
acceptance testing and then you launch and this is just generally
4:21
launch and this is just generally
4:21
launch and this is just generally accepted oh it's going to take 12
4:24
accepted oh it's going to take 12
4:24
accepted oh it's going to take 12 months um what I have found is that if
4:28
months um what I have found is that if
4:28
months um what I have found is that if you plan
4:30
you plan
4:30
you plan then you can make it an eight-month
4:31
then you can make it an eight-month
4:31
then you can make it an eight-month project so what would have been 12
4:33
project so what would have been 12
4:33
project so what would have been 12 months is now eight
4:35
months is now eight
4:35
months is now eight months and that eight months can often
4:38
months and that eight months can often
4:38
months and that eight months can often even start
4:39
even start
4:39
even start earlier and really it because planning
4:43
earlier and really it because planning
4:43
earlier and really it because planning happens as part of requirements and it's
4:46
happens as part of requirements and it's
4:46
happens as part of requirements and it's really about a meshing of the business
4:48
really about a meshing of the business
4:48
really about a meshing of the business and the technology side at the same
4:51
and the technology side at the same
4:51
and the technology side at the same time so what goes into the planning
4:54
time so what goes into the planning
4:54
time so what goes into the planning phase um the first is understanding the
4:57
phase um the first is understanding the
4:57
phase um the first is understanding the requirements um too many times the
5:00
requirements um too many times the
5:00
requirements um too many times the requirements are just kind of handed
5:01
requirements are just kind of handed
5:02
requirements are just kind of handed over um and like I I mean for anyone
5:06
over um and like I I mean for anyone
5:06
over um and like I I mean for anyone who's old enough to remember the
5:07
who's old enough to remember the
5:07
who's old enough to remember the waterfall model and getting you know
5:09
waterfall model and getting you know
5:10
waterfall model and getting you know their BRD documents and and and just
5:12
their BRD documents and and and just
5:12
their BRD documents and and and just these massive 50 page things um as as
5:17
these massive 50 page things um as as
5:17
these massive 50 page things um as as much as it was great to get
5:18
much as it was great to get
5:18
much as it was great to get specifications a lot of times developers
5:22
specifications a lot of times developers
5:22
specifications a lot of times developers the ones writing the code didn't
5:24
the ones writing the code didn't
5:24
the ones writing the code didn't understand what problem they were trying
5:25
understand what problem they were trying
5:25
understand what problem they were trying to solve so you would get people who
5:29
to solve so you would get people who
5:29
to solve so you would get people who would spend time developing something
5:31
would spend time developing something
5:31
would spend time developing something that that needed to be reworked so
5:34
that that needed to be reworked so
5:34
that that needed to be reworked so really day one is just being part of the
5:38
really day one is just being part of the
5:38
really day one is just being part of the business talking with the business to
5:40
business talking with the business to
5:40
business talking with the business to understand the
5:43
understand the
5:43
understand the requirements um the second one and this
5:45
requirements um the second one and this
5:45
requirements um the second one and this is important evaluate many code
5:47
is important evaluate many code
5:47
is important evaluate many code framework
5:48
framework
5:48
framework options um often times what people say
5:53
options um often times what people say
5:53
options um often times what people say well I've I've laid code I've started
5:54
well I've I've laid code I've started
5:54
well I've I've laid code I've started writing code I'm just going to go so you
5:57
writing code I'm just going to go so you
5:57
writing code I'm just going to go so you pick react as a front-end framework and
5:59
pick react as a front-end framework and
5:59
pick react as a front-end framework and you run with it you pick you know
6:01
you run with it you pick you know
6:01
you run with it you pick you know Tailwind you pick whatever
6:04
Tailwind you pick whatever
6:04
Tailwind you pick whatever and at the end of the day every
6:07
and at the end of the day every
6:07
and at the end of the day every framework has positives and every
6:08
framework has positives and every
6:08
framework has positives and every framework has negatives and you need to
6:11
framework has negatives and you need to
6:11
framework has negatives and you need to understand what framework will suit your
6:14
understand what framework will suit your
6:14
understand what framework will suit your full needs best and the only way to do
6:17
full needs best and the only way to do
6:17
full needs best and the only way to do that is to evaluate more than one so if
6:19
that is to evaluate more than one so if
6:19
that is to evaluate more than one so if you're spinning up a react project you
6:21
you're spinning up a react project you
6:21
you're spinning up a react project you should also be spinning up a next
6:23
should also be spinning up a next
6:23
should also be spinning up a next project maybe spin up a view project you
6:25
project maybe spin up a view project you
6:25
project maybe spin up a view project you know your developers during the time
6:28
know your developers during the time
6:28
know your developers during the time where require ments are being gathered
6:31
where require ments are being gathered
6:31
where require ments are being gathered should be playing with different
6:32
should be playing with different
6:32
should be playing with different Frameworks trying to understand in a
6:34
Frameworks trying to understand in a
6:34
Frameworks trying to understand in a bigger picture what's the right solution
6:37
bigger picture what's the right solution
6:37
bigger picture what's the right solution and what's the right
6:40
framework having dialogue with your
6:43
framework having dialogue with your
6:43
framework having dialogue with your stakeholders so the the technology team
6:46
stakeholders so the the technology team
6:46
stakeholders so the the technology team needs to be a stakeholder they need to
6:49
needs to be a stakeholder they need to
6:49
needs to be a stakeholder they need to have investment in the success of the
6:51
have investment in the success of the
6:51
have investment in the success of the application but they also need to be
6:53
application but they also need to be
6:53
application but they also need to be available early on and part of that
6:58
discussion now this is this is what's
7:01
discussion now this is this is what's
7:01
discussion now this is this is what's called the this is also known as the
7:02
called the this is also known as the
7:02
called the this is also known as the expert trap so how many times have we
7:06
expert trap so how many times have we
7:06
expert trap so how many times have we been asked a
7:07
been asked a
7:07
been asked a question and the first option we come up
7:10
question and the first option we come up
7:10
question and the first option we come up with all of a sudden becomes the best
7:12
with all of a sudden becomes the best
7:12
with all of a sudden becomes the best option and the only solution and we run
7:14
option and the only solution and we run
7:14
option and the only solution and we run with that um you know it's like hey we
7:17
with that um you know it's like hey we
7:17
with that um you know it's like hey we need a database MySQL let's just go with
7:20
need a database MySQL let's just go with
7:20
need a database MySQL let's just go with mySQL so you know all of a sudden now
7:22
mySQL so you know all of a sudden now
7:22
mySQL so you know all of a sudden now we're using my SQL but we didn't stop
7:25
we're using my SQL but we didn't stop
7:25
we're using my SQL but we didn't stop and say do we need a SQL database do we
7:28
and say do we need a SQL database do we
7:28
and say do we need a SQL database do we actually need a vector database do we
7:30
actually need a vector database do we
7:30
actually need a vector database do we need something that's more nosql do we
7:32
need something that's more nosql do we
7:32
need something that's more nosql do we need so just because we have an option
7:35
need so just because we have an option
7:35
need so just because we have an option that sounds good don't stop and actually
7:38
that sounds good don't stop and actually
7:39
that sounds good don't stop and actually spend more time determining if that if
7:41
spend more time determining if that if
7:41
spend more time determining if that if that answer is the right answer or the
7:44
that answer is the right answer or the
7:44
that answer is the right answer or the best answer because there's multiple
7:45
best answer because there's multiple
7:46
best answer because there's multiple right
7:47
right
7:47
right answers um and the last thing that
7:49
answers um and the last thing that
7:49
answers um and the last thing that happens should be planning in planning
7:51
happens should be planning in planning
7:51
happens should be planning in planning should be out of the-box
7:53
should be out of the-box
7:53
should be out of the-box solutions uh as coders we like to solve
7:57
solutions uh as coders we like to solve
7:57
solutions uh as coders we like to solve problems with code um there's a a phrase
8:00
problems with code um there's a a phrase
8:00
problems with code um there's a a phrase I use often when all you have is a
8:02
I use often when all you have is a
8:02
I use often when all you have is a hammer everything looks like a nail and
8:05
hammer everything looks like a nail and
8:05
hammer everything looks like a nail and if you try and nail a screw into a piece
8:07
if you try and nail a screw into a piece
8:07
if you try and nail a screw into a piece of wood it doesn't work very well it
8:10
of wood it doesn't work very well it
8:11
of wood it doesn't work very well it works you can get the screw in there
8:12
works you can get the screw in there
8:12
works you can get the screw in there trust me as a failed Carpenter I can
8:15
trust me as a failed Carpenter I can
8:15
trust me as a failed Carpenter I can tell you you can do it um but so much
8:20
tell you you can do it um but so much
8:20
tell you you can do it um but so much development time can be saved by buying
8:23
development time can be saved by buying
8:23
development time can be saved by buying software and I I know sometimes like the
8:25
software and I I know sometimes like the
8:25
software and I I know sometimes like the business like I'll I'll be in a
8:27
business like I'll I'll be in a
8:27
business like I'll I'll be in a situation where I'll say there's piece
8:29
situation where I'll say there's piece
8:29
situation where I'll say there's piece of software that'll cost us $1,000 and
8:32
of software that'll cost us $1,000 and
8:32
of software that'll cost us $1,000 and they'll go whoa that's crazy and I'm
8:36
they'll go whoa that's crazy and I'm
8:36
they'll go whoa that's crazy and I'm like our developers cost more than $
8:38
like our developers cost more than $
8:38
like our developers cost more than $ thousand dollar a week so what I'm
8:40
thousand dollar a week so what I'm
8:40
thousand dollar a week so what I'm saying here is you can spend a whole
8:42
saying here is you can spend a whole
8:43
saying here is you can spend a whole month developing something or we can buy
8:45
month developing something or we can buy
8:45
month developing something or we can buy it for, and all a sudden it's like well
8:48
it for, and all a sudden it's like well
8:48
it for, and all a sudden it's like well yeah okay that's a whole lot cheaper so
8:51
yeah okay that's a whole lot cheaper so
8:51
yeah okay that's a whole lot cheaper so out of the box Solutions if they work
8:54
out of the box Solutions if they work
8:54
out of the box Solutions if they work are a huge way to save time and
8:57
are a huge way to save time and
8:57
are a huge way to save time and money uh and then understanding roughly
9:00
money uh and then understanding roughly
9:00
money uh and then understanding roughly what you're expecting now tomorrow and
9:04
what you're expecting now tomorrow and
9:04
what you're expecting now tomorrow and what requirements are your wish list
9:08
items all right you are not paid to code
9:11
items all right you are not paid to code
9:11
items all right you are not paid to code you are paid to solve
9:13
you are paid to solve
9:13
you are paid to solve problems um so I'm going to repeat
9:16
problems um so I'm going to repeat
9:16
problems um so I'm going to repeat that if the
9:19
that if the
9:19
that if the Technology Group if you're expected to
9:22
Technology Group if you're expected to
9:22
Technology Group if you're expected to solve a Content management
9:25
solve a Content management
9:25
solve a Content management system you are not paid to write
9:28
system you are not paid to write
9:28
system you are not paid to write thousands of Lies of code you're paid to
9:32
thousands of Lies of code you're paid to
9:32
thousands of Lies of code you're paid to solve the content management system so
9:34
solve the content management system so
9:34
solve the content management system so if WordPress solves your
9:36
if WordPress solves your
9:36
if WordPress solves your needs you are of more value choosing
9:40
needs you are of more value choosing
9:40
needs you are of more value choosing WordPress and maintaining that than you
9:43
WordPress and maintaining that than you
9:43
WordPress and maintaining that than you are writing your own content management
9:44
are writing your own content management
9:44
are writing your own content management system assuming that
9:47
system assuming that
9:47
system assuming that works all right 90% of apps are over
9:50
works all right 90% of apps are over
9:50
works all right 90% of apps are over engineered
9:52
engineered
9:52
engineered um if you ever work on an app that you
9:55
um if you ever work on an app that you
9:55
um if you ever work on an app that you feel like you know what this is this is
9:58
feel like you know what this is this is
9:58
feel like you know what this is this is right size it's perfect there's nothing
10:00
right size it's perfect there's nothing
10:00
right size it's perfect there's nothing in here that's that's unnecessary I'd be
10:03
in here that's that's unnecessary I'd be
10:03
in here that's that's unnecessary I'd be surprised um it shocks me that I mean
10:07
surprised um it shocks me that I mean
10:07
surprised um it shocks me that I mean any front-end application that uses
10:09
any front-end application that uses
10:09
any front-end application that uses react uses a bunch of different uh node
10:13
react uses a bunch of different uh node
10:13
react uses a bunch of different uh node modules uh and then you add it all up
10:16
modules uh and then you add it all up
10:16
modules uh and then you add it all up and it's 600 megabytes for a two-page
10:19
and it's 600 megabytes for a two-page
10:19
and it's 600 megabytes for a two-page application it's over
10:22
application it's over
10:22
application it's over engineered um when you over engineer
10:25
engineered um when you over engineer
10:25
engineered um when you over engineer code you frustrate yourself you distract
10:28
code you frustrate yourself you distract
10:28
code you frustrate yourself you distract future you from from solving new
10:30
future you from from solving new
10:30
future you from from solving new problems you increase the difficulty of
10:32
problems you increase the difficulty of
10:32
problems you increase the difficulty of new development for future U and you're
10:35
new development for future U and you're
10:35
new development for future U and you're embarrassing future U when you have to
10:38
embarrassing future U when you have to
10:38
embarrassing future U when you have to explain why you have to rewrite code
10:40
explain why you have to rewrite code
10:40
explain why you have to rewrite code that you wrote in the past so love
10:43
that you wrote in the past so love
10:43
that you wrote in the past so love yourself current you and future
10:46
yourself current you and future
10:46
yourself current you and future you all right so what are the phases of
10:50
you all right so what are the phases of
10:50
you all right so what are the phases of development
10:52
development
10:52
development so when I think of
10:55
so when I think of
10:55
so when I think of projects um I mean they really break
10:57
projects um I mean they really break
10:57
projects um I mean they really break down into three
11:00
down into three
11:00
down into three main three main parts of a life cycle
11:02
main three main parts of a life cycle
11:02
main three main parts of a life cycle the first is the proof of concept so if
11:05
the first is the proof of concept so if
11:05
the first is the proof of concept so if you work in a startup um pretty much
11:08
you work in a startup um pretty much
11:08
you work in a startup um pretty much everything you're doing is proof of
11:09
everything you're doing is proof of
11:09
everything you're doing is proof of concept and a proof of concept is a
11:12
concept and a proof of concept is a
11:12
concept and a proof of concept is a great way to show that something works
11:14
great way to show that something works
11:14
great way to show that something works so a lot of people are building proof of
11:16
so a lot of people are building proof of
11:16
so a lot of people are building proof of Concepts on top of things like open AI
11:18
Concepts on top of things like open AI
11:18
Concepts on top of things like open AI it's like all right I have something it
11:20
it's like all right I have something it
11:20
it's like all right I have something it throws a prompt in there gets the
11:21
throws a prompt in there gets the
11:21
throws a prompt in there gets the response and displays it and it's a
11:24
response and displays it and it's a
11:24
response and displays it and it's a great proof of concept but it doesn't
11:26
great proof of concept but it doesn't
11:26
great proof of concept but it doesn't scale it often you know can't can't go
11:29
scale it often you know can't can't go
11:29
scale it often you know can't can't go beyond what chat GPT can do and you
11:32
beyond what chat GPT can do and you
11:32
beyond what chat GPT can do and you really you know there's there's
11:33
really you know there's there's
11:33
really you know there's there's limitations so proof of concept is
11:35
limitations so proof of concept is
11:35
limitations so proof of concept is usually small Limited in scope it does
11:38
usually small Limited in scope it does
11:38
usually small Limited in scope it does one thing it does it really well and
11:41
one thing it does it really well and
11:41
one thing it does it really well and it's just meant to see is this worth
11:42
it's just meant to see is this worth
11:42
it's just meant to see is this worth developing
11:44
developing
11:44
developing more where I see a lot of businesses
11:48
more where I see a lot of businesses
11:48
more where I see a lot of businesses make bad decisions is then they go into
11:50
make bad decisions is then they go into
11:50
make bad decisions is then they go into phase two which is boil the ocean and
11:53
phase two which is boil the ocean and
11:53
phase two which is boil the ocean and that's when you take a proof of concept
11:56
that's when you take a proof of concept
11:56
that's when you take a proof of concept and you try and blow it up so it's like
11:59
and you try and blow it up so it's like
11:59
and you try and blow it up so it's like okay great this does a but we want b c d
12:02
okay great this does a but we want b c d
12:03
okay great this does a but we want b c d and e and while we're at it it's got to
12:04
and e and while we're at it it's got to
12:05
and e and while we're at it it's got to do f and g it's got to handle a million
12:07
do f and g it's got to handle a million
12:07
do f and g it's got to handle a million concurrent users it's got to um have
12:10
concurrent users it's got to um have
12:10
concurrent users it's got to um have this and this and this and this is a you
12:12
this and this and this and this is a you
12:12
this and this and this and this is a you know chat GPT is new so it's got to
12:14
know chat GPT is new so it's got to
12:14
know chat GPT is new so it's got to integrate chat GPT and it's also has to
12:16
integrate chat GPT and it's also has to
12:16
integrate chat GPT and it's also has to have its own chat bot but then it's got
12:18
have its own chat bot but then it's got
12:18
have its own chat bot but then it's got to and your requirements get out of
12:20
to and your requirements get out of
12:20
to and your requirements get out of control and all of that happens and
12:24
control and all of that happens and
12:24
control and all of that happens and oftentimes especially if your proof of
12:26
oftentimes especially if your proof of
12:26
oftentimes especially if your proof of concept is good that actually gets
12:28
concept is good that actually gets
12:28
concept is good that actually gets approved
12:29
approved
12:29
approved and you have this crazy idea and then
12:32
and you have this crazy idea and then
12:32
and you have this crazy idea and then the reality sets in of how much work
12:34
the reality sets in of how much work
12:34
the reality sets in of how much work it's going to be once you've got a few
12:36
it's going to be once you've got a few
12:36
it's going to be once you've got a few launches you know behind you and then
12:39
launches you know behind you and then
12:39
launches you know behind you and then you like wait we have a codebase that's
12:41
you like wait we have a codebase that's
12:41
you like wait we have a codebase that's untenable it's there's it's just a mess
12:44
untenable it's there's it's just a mess
12:44
untenable it's there's it's just a mess it's spaghetti so we have to do it all
12:46
it's spaghetti so we have to do it all
12:46
it's spaghetti so we have to do it all over again and then you go into phase
12:48
over again and then you go into phase
12:48
over again and then you go into phase three which is actually what your
12:49
three which is actually what your
12:49
three which is actually what your customer wants and in most cases that's
12:52
customer wants and in most cases that's
12:52
customer wants and in most cases that's a limited feature set limited
12:55
a limited feature set limited
12:55
a limited feature set limited requirements and hopefully a simple
12:58
requirements and hopefully a simple
12:58
requirements and hopefully a simple enough architecture you can keep going
13:00
enough architecture you can keep going
13:00
enough architecture you can keep going so by planning the idea is to skip phase
13:05
so by planning the idea is to skip phase
13:05
so by planning the idea is to skip phase two Al together and go right from proof
13:07
two Al together and go right from proof
13:07
two Al together and go right from proof of concept into what your customer
13:12
of concept into what your customer
13:12
of concept into what your customer wants all right so some stories so um a
13:16
wants all right so some stories so um a
13:16
wants all right so some stories so um a startup that I worked at several years
13:18
startup that I worked at several years
13:18
startup that I worked at several years ago uh 15ish years ago so we started
13:22
ago uh 15ish years ago so we started
13:22
ago uh 15ish years ago so we started with a proof of concept uh we took a
13:25
with a proof of concept uh we took a
13:25
with a proof of concept uh we took a pre-built software it was I I think
13:28
pre-built software it was I I think
13:28
pre-built software it was I I think roughly $500 template and it was a DVD
13:30
roughly $500 template and it was a DVD
13:30
roughly $500 template and it was a DVD rental service so we were going to be a
13:33
rental service so we were going to be a
13:33
rental service so we were going to be a book rental service so we were kind of
13:34
book rental service so we were kind of
13:34
book rental service so we were kind of like Netflix but for
13:36
like Netflix but for
13:36
like Netflix but for books and uh the DVD rental it worked
13:41
books and uh the DVD rental it worked
13:41
books and uh the DVD rental it worked reasonably well we were able to you know
13:43
reasonably well we were able to you know
13:43
reasonably well we were able to you know make it work with isbns as opposed to uh
13:46
make it work with isbns as opposed to uh
13:46
make it work with isbns as opposed to uh you know the eens on a DVD um we were
13:49
you know the eens on a DVD um we were
13:49
you know the eens on a DVD um we were able to get it with to work with book
13:51
able to get it with to work with book
13:52
able to get it with to work with book data um so ultimately we saved a whole
13:55
data um so ultimately we saved a whole
13:55
data um so ultimately we saved a whole bunch of time because we used something
13:57
bunch of time because we used something
13:57
bunch of time because we used something that existed
13:59
that existed
13:59
that existed it took one developer and two interns
14:01
it took one developer and two interns
14:01
it took one developer and two interns about three months to customize this and
14:03
about three months to customize this and
14:03
about three months to customize this and launch this so for a proof of concept
14:05
launch this so for a proof of concept
14:05
launch this so for a proof of concept that was really solid um but we had
14:09
that was really solid um but we had
14:09
that was really solid um but we had performance issues we couldn't grow we
14:11
performance issues we couldn't grow we
14:11
performance issues we couldn't grow we couldn't extend it um there was some
14:14
couldn't extend it um there was some
14:14
couldn't extend it um there was some security concerns so at the end of the
14:17
security concerns so at the end of the
14:17
security concerns so at the end of the day as a proof concept it was great but
14:18
day as a proof concept it was great but
14:18
day as a proof concept it was great but we had to rewrite it um so when we
14:22
we had to rewrite it um so when we
14:22
we had to rewrite it um so when we rewrote it there was a fancy new PHP
14:24
rewrote it there was a fancy new PHP
14:24
rewrote it there was a fancy new PHP framework that had just come out so we
14:26
framework that had just come out so we
14:26
framework that had just come out so we decided to apply that um because the
14:29
decided to apply that um because the
14:29
decided to apply that um because the lead designer at the time you know
14:31
lead designer at the time you know
14:31
lead designer at the time you know wanted to do the bleeding edge of
14:33
wanted to do the bleeding edge of
14:33
wanted to do the bleeding edge of everything um we also tried to do
14:36
everything um we also tried to do
14:36
everything um we also tried to do wehouse management Inventory management
14:39
wehouse management Inventory management
14:39
wehouse management Inventory management purchasing system we tried to do
14:41
purchasing system we tried to do
14:41
purchasing system we tried to do everything so really we entered boil the
14:43
everything so really we entered boil the
14:43
everything so really we entered boil the ocean
14:45
ocean
14:45
ocean um the project was good but it got over
14:49
um the project was good but it got over
14:49
um the project was good but it got over engineered and we very quickly found
14:52
engineered and we very quickly found
14:52
engineered and we very quickly found that that PHP framework ended up biting
14:55
that that PHP framework ended up biting
14:55
that that PHP framework ended up biting Us in the butt it was not new enough or
14:57
Us in the butt it was not new enough or
14:57
Us in the butt it was not new enough or it was not uh out long enough enough for
14:59
it was not uh out long enough enough for
14:59
it was not uh out long enough enough for it to kind of get through its issues um
15:03
it to kind of get through its issues um
15:03
it to kind of get through its issues um it was going to take too long to report
15:04
it was going to take too long to report
15:04
it was going to take too long to report every bug it just it wasn't right and
15:09
every bug it just it wasn't right and
15:09
every bug it just it wasn't right and about a year in nine months in we had to
15:12
about a year in nine months in we had to
15:12
about a year in nine months in we had to rewrite it from scratch and this time we
15:15
rewrite it from scratch and this time we
15:15
rewrite it from scratch and this time we just wrote it in in straight PHP we
15:17
just wrote it in in straight PHP we
15:18
just wrote it in in straight PHP we wrote a very small lightweight framework
15:20
wrote a very small lightweight framework
15:20
wrote a very small lightweight framework on top of it uh that we wrote in house
15:23
on top of it uh that we wrote in house
15:23
on top of it uh that we wrote in house that is super simple an intern could
15:26
that is super simple an intern could
15:26
that is super simple an intern could learn it a day or two um and it it made
15:29
learn it a day or two um and it it made
15:29
learn it a day or two um and it it made our development simple and we did manage
15:32
our development simple and we did manage
15:32
our development simple and we did manage to get warehouse and inventory
15:33
to get warehouse and inventory
15:33
to get warehouse and inventory management in and we got a few other
15:35
management in and we got a few other
15:35
management in and we got a few other things but we did it on a much smaller
15:37
things but we did it on a much smaller
15:37
things but we did it on a much smaller scale because we weren't
15:39
scale because we weren't
15:39
scale because we weren't Netflix and uh it took a little longer
15:43
Netflix and uh it took a little longer
15:43
Netflix and uh it took a little longer to develop but we had a much more robust
15:45
to develop but we had a much more robust
15:45
to develop but we had a much more robust system that the reality was it actually
15:48
system that the reality was it actually
15:48
system that the reality was it actually ran itself it handled customer service
15:51
ran itself it handled customer service
15:51
ran itself it handled customer service almost completely automatically it
15:52
almost completely automatically it
15:52
almost completely automatically it handled purchasing almost completely
15:54
handled purchasing almost completely
15:54
handled purchasing almost completely automatically so we were able to take 12
15:58
automatically so we were able to take 12
15:58
automatically so we were able to take 12 work and bring it down to
16:00
work and bring it down to
16:00
work and bring it down to seven um so it you know it just really
16:04
seven um so it you know it just really
16:04
seven um so it you know it just really was all about planning in that
16:07
was all about planning in that
16:07
was all about planning in that stage um so I I talked about what we
16:09
stage um so I I talked about what we
16:10
stage um so I I talked about what we learned so the Frameworks were just not
16:11
learned so the Frameworks were just not
16:11
learned so the Frameworks were just not there um you have to really design for
16:14
there um you have to really design for
16:14
there um you have to really design for what you need oftentimes we get so
16:16
what you need oftentimes we get so
16:16
what you need oftentimes we get so excited we want to do everything we want
16:18
excited we want to do everything we want
16:18
excited we want to do everything we want to boil the ocean and you really gota
16:20
to boil the ocean and you really gota
16:20
to boil the ocean and you really gota got to slim that down uh we cut a lot of
16:23
got to slim that down uh we cut a lot of
16:23
got to slim that down uh we cut a lot of Corners in the beginning and every
16:25
Corners in the beginning and every
16:25
Corners in the beginning and every corner we cut bid us uh and I'm just
16:28
corner we cut bid us uh and I'm just
16:28
corner we cut bid us uh and I'm just going to keep reiterating plan plan
16:32
going to keep reiterating plan plan
16:32
going to keep reiterating plan plan plan all right
16:34
plan all right
16:34
plan all right so to answer hopefully a question that
16:37
so to answer hopefully a question that
16:37
so to answer hopefully a question that might be in your mind where do
16:38
might be in your mind where do
16:38
might be in your mind where do developers and requirements sit um when
16:41
developers and requirements sit um when
16:41
developers and requirements sit um when you work for a startup and you've got
16:43
you work for a startup and you've got
16:43
you work for a startup and you've got eight people in the company you're all
16:45
eight people in the company you're all
16:45
eight people in the company you're all talking to each other you all know
16:46
talking to each other you all know
16:46
talking to each other you all know what's going on um you're probably kind
16:49
what's going on um you're probably kind
16:49
what's going on um you're probably kind of already part of it if you're a
16:51
of already part of it if you're a
16:51
of already part of it if you're a oneperson team you're everything this
16:52
oneperson team you're everything this
16:52
oneperson team you're everything this isn't for you when you work for a mid or
16:55
isn't for you when you work for a mid or
16:55
isn't for you when you work for a mid or large siiz company there can be a
16:57
large siiz company there can be a
16:57
large siiz company there can be a business side and a Tech side that don't
17:00
business side and a Tech side that don't
17:00
business side and a Tech side that don't talk to each other very well the
17:01
talk to each other very well the
17:01
talk to each other very well the business side sends requirements Tech
17:04
business side sends requirements Tech
17:04
business side sends requirements Tech side writes the software um so what I
17:09
side writes the software um so what I
17:09
side writes the software um so what I have found works really well is you have
17:12
have found works really well is you have
17:12
have found works really well is you have to break those silos down you have to
17:15
to break those silos down you have to
17:15
to break those silos down you have to bring the tech side over to the business
17:18
bring the tech side over to the business
17:18
bring the tech side over to the business and you have to make sure they
17:19
and you have to make sure they
17:19
and you have to make sure they understand the kpis kpi is a key
17:21
understand the kpis kpi is a key
17:21
understand the kpis kpi is a key performance indicator so let's say
17:25
performance indicator so let's say
17:25
performance indicator so let's say you're an e-commerce website your
17:27
you're an e-commerce website your
17:27
you're an e-commerce website your conversion rate your your add to cart
17:29
conversion rate your your add to cart
17:29
conversion rate your your add to cart rate is going to be a key performance
17:31
rate is going to be a key performance
17:31
rate is going to be a key performance indicator
17:33
indicator
17:33
indicator your um checkout completion rate so the
17:37
your um checkout completion rate so the
17:37
your um checkout completion rate so the from add to cart to checkout so there's
17:40
from add to cart to checkout so there's
17:40
from add to cart to checkout so there's going to be certain items that are going
17:42
going to be certain items that are going
17:42
going to be certain items that are going to be really important and knowing what
17:44
to be really important and knowing what
17:44
to be really important and knowing what those are will help you as a developer
17:46
those are will help you as a developer
17:46
those are will help you as a developer because if you know you need to make
17:49
because if you know you need to make
17:49
because if you know you need to make something easier then when you're
17:51
something easier then when you're
17:51
something easier then when you're developing you're going to think does
17:53
developing you're going to think does
17:53
developing you're going to think does this make it
17:54
this make it
17:54
this make it easier um as because sometimes
17:57
easier um as because sometimes
17:57
easier um as because sometimes requirements get confusing
17:59
requirements get confusing
17:59
requirements get confusing you have to know who your personas are
18:01
you have to know who your personas are
18:01
you have to know who your personas are you have to know who is using your site
18:03
you have to know who is using your site
18:03
you have to know who is using your site what is their motivations what do they
18:06
what is their motivations what do they
18:06
what is their motivations what do they want out of your site and Are you
18:08
want out of your site and Are you
18:08
want out of your site and Are you delivering that you may not be a design
18:10
delivering that you may not be a design
18:10
delivering that you may not be a design expert and know if a green button or a
18:12
expert and know if a green button or a
18:12
expert and know if a green button or a blue button or a red button is going to
18:14
blue button or a red button is going to
18:14
blue button or a red button is going to be more successful but if you know that
18:16
be more successful but if you know that
18:16
be more successful but if you know that your target audience is over the age of
18:18
your target audience is over the age of
18:18
your target audience is over the age of 60 and you're using eighto font you
18:21
60 and you're using eighto font you
18:21
60 and you're using eighto font you should know that that it's not going to
18:24
should know that that it's not going to
18:24
should know that that it's not going to work um and as I saying developers must
18:28
work um and as I saying developers must
18:28
work um and as I saying developers must the key stakeholders the developers need
18:31
the key stakeholders the developers need
18:31
the key stakeholders the developers need to be part of the requirements process
18:33
to be part of the requirements process
18:33
to be part of the requirements process even if it's just to talk through what
18:36
even if it's just to talk through what
18:36
even if it's just to talk through what possible solutions are and I think
18:38
possible solutions are and I think
18:39
possible solutions are and I think what's critical there
18:41
what's critical there
18:41
what's critical there is oftentimes the business will want
18:46
is oftentimes the business will want
18:46
is oftentimes the business will want let's say three features so let's say a
18:48
let's say three features so let's say a
18:48
let's say three features so let's say a search engine so we want a search engine
18:51
search engine so we want a search engine
18:51
search engine so we want a search engine and let's say so I work in the book
18:53
and let's say so I work in the book
18:53
and let's say so I work in the book world we want to search and we want to
18:54
world we want to search and we want to
18:54
world we want to search and we want to be able to search on title we want to be
18:57
be able to search on title we want to be
18:57
be able to search on title we want to be able to search on author and we want to
19:00
able to search on author and we want to
19:00
able to search on author and we want to be able to search on
19:03
be able to search on
19:03
be able to search on themes so you talk through this and
19:05
themes so you talk through this and
19:05
themes so you talk through this and you're like well title super easy
19:07
you're like well title super easy
19:07
you're like well title super easy because we have the data it exists
19:09
because we have the data it exists
19:09
because we have the data it exists authors super easy we have that data we
19:12
authors super easy we have that data we
19:12
authors super easy we have that data we can do it themes I don't know what you
19:15
can do it themes I don't know what you
19:15
can do it themes I don't know what you mean by a
19:16
mean by a
19:17
mean by a theme and so now you can have a
19:19
theme and so now you can have a
19:19
theme and so now you can have a discussion and you can talk to the
19:20
discussion and you can talk to the
19:20
discussion and you can talk to the business about what data does exist what
19:22
business about what data does exist what
19:22
business about what data does exist what data doesn't exist how much time it will
19:25
data doesn't exist how much time it will
19:25
data doesn't exist how much time it will take because if you're there early in
19:28
take because if you're there early in
19:28
take because if you're there early in the
19:29
the
19:29
the process you have a chance to say well
19:31
process you have a chance to say well
19:31
process you have a chance to say well title author is easy it'll take us less
19:33
title author is easy it'll take us less
19:33
title author is easy it'll take us less than a week theme is going to be hard
19:35
than a week theme is going to be hard
19:35
than a week theme is going to be hard it's going to take us a month and the
19:37
it's going to take us a month and the
19:37
it's going to take us a month and the business can say you know what you're
19:38
business can say you know what you're
19:38
business can say you know what you're right theme is not worth worth one month
19:41
right theme is not worth worth one month
19:41
right theme is not worth worth one month of effort so you've cut an entire month
19:44
of effort so you've cut an entire month
19:44
of effort so you've cut an entire month off of your Dev timeline because you had
19:46
off of your Dev timeline because you had
19:46
off of your Dev timeline because you had a single
19:47
a single
19:47
a single conversation or the conversation could
19:49
conversation or the conversation could
19:49
conversation or the conversation could go here's the data we do have so instead
19:52
go here's the data we do have so instead
19:52
go here's the data we do have so instead of theme can we change it to
19:54
of theme can we change it to
19:55
of theme can we change it to keywords and they're like yeah keywords
19:57
keywords and they're like yeah keywords
19:57
keywords and they're like yeah keywords is great you're having that discussion
19:59
is great you're having that discussion
19:59
is great you're having that discussion up front as opposed to once the business
20:02
up front as opposed to once the business
20:02
up front as opposed to once the business has committed and and like here's our
20:04
has committed and and like here's our
20:04
has committed and and like here's our timeline here's what we want and then
20:06
timeline here's what we want and then
20:06
timeline here's what we want and then you come back and say well it's going to
20:07
you come back and say well it's going to
20:07
you come back and say well it's going to take a month and two weeks you know and
20:09
take a month and two weeks you know and
20:09
take a month and two weeks you know and it's just so you're you're extending
20:12
it's just so you're you're extending
20:12
it's just so you're you're extending your timeline when it may not be
20:15
necessary all right so another project
20:18
necessary all right so another project
20:18
necessary all right so another project that I've you know went on that had a
20:20
that I've you know went on that had a
20:20
that I've you know went on that had a similar type uh process was phase a was
20:23
similar type uh process was phase a was
20:23
similar type uh process was phase a was proof of concept it was a single
20:25
proof of concept it was a single
20:25
proof of concept it was a single developer this was uh a startup that I
20:28
developer this was uh a startup that I
20:28
developer this was uh a startup that I had
20:29
had
20:29
had and um every day was a new requirement
20:33
and um every day was a new requirement
20:33
and um every day was a new requirement so every day was just heaping on more
20:35
so every day was just heaping on more
20:35
so every day was just heaping on more and more and more code we got through
20:37
and more and more code we got through
20:37
and more and more code we got through two years of development it got us
20:39
two years of development it got us
20:39
two years of development it got us through our two rounds of funding it got
20:42
through our two rounds of funding it got
20:42
through our two rounds of funding it got us to
20:43
us to
20:43
us to acquisition um but I I'm proud of what
20:47
acquisition um but I I'm proud of what
20:47
acquisition um but I I'm proud of what the code did but I'm not proud of the
20:49
the code did but I'm not proud of the
20:49
the code did but I'm not proud of the code um the code is a mess I like to say
20:51
code um the code is a mess I like to say
20:51
code um the code is a mess I like to say it was held together with duct tape and
20:53
it was held together with duct tape and
20:53
it was held together with duct tape and safety pins um it it was not scalable
20:58
safety pins um it it was not scalable
20:58
safety pins um it it was not scalable but it did what it was supposed to do so
21:00
but it did what it was supposed to do so
21:00
but it did what it was supposed to do so once we got
21:02
once we got
21:02
once we got acquired they had a really tight
21:06
acquired they had a really tight
21:06
acquired they had a really tight deadline to try and get us to a certain
21:08
deadline to try and get us to a certain
21:08
deadline to try and get us to a certain Revenue number so instead of saying well
21:11
Revenue number so instead of saying well
21:11
Revenue number so instead of saying well now you have this proof of concept that
21:13
now you have this proof of concept that
21:13
now you have this proof of concept that we need to throw away and rewrite they
21:16
we need to throw away and rewrite they
21:16
we need to throw away and rewrite they said you have to just keep adding on to
21:18
said you have to just keep adding on to
21:18
said you have to just keep adding on to it and you have to get all of the legal
21:21
it and you have to get all of the legal
21:21
it and you have to get all of the legal requirements all of you know everything
21:23
requirements all of you know everything
21:23
requirements all of you know everything else that we had neglected so we just
21:26
else that we had neglected so we just
21:26
else that we had neglected so we just kept building and it became a behemoth
21:29
kept building and it became a behemoth
21:29
kept building and it became a behemoth so the boil the ocean stage was just
21:32
so the boil the ocean stage was just
21:32
so the boil the ocean stage was just slapping on code after code after code
21:36
slapping on code after code after code
21:36
slapping on code after code after code and um it became unmaintainable and it
21:40
and um it became unmaintainable and it
21:40
and um it became unmaintainable and it you know it was we re you were hitting
21:42
you know it was we re you were hitting
21:42
you know it was we re you were hitting lots of performance issues because
21:44
lots of performance issues because
21:44
lots of performance issues because scalability wasn't in mind it just it it
21:47
scalability wasn't in mind it just it it
21:47
scalability wasn't in mind it just it it it hit that point where we had to
21:48
it hit that point where we had to
21:48
it hit that point where we had to rewrite everything um now there was a
21:52
rewrite everything um now there was a
21:52
rewrite everything um now there was a lot of decisions made in that rewrite I
21:54
lot of decisions made in that rewrite I
21:54
lot of decisions made in that rewrite I bring this up because I think it's a
21:55
bring this up because I think it's a
21:55
bring this up because I think it's a great Learning lesson so
21:58
great Learning lesson so
21:58
great Learning lesson so we decided to go and be try try and be
22:02
we decided to go and be try try and be
22:02
we decided to go and be try try and be leaders and go into a microservices
22:05
leaders and go into a microservices
22:05
leaders and go into a microservices architecture so the design here and
22:08
architecture so the design here and
22:08
architecture so the design here and where we failed to plan enough
22:11
where we failed to plan enough
22:11
where we failed to plan enough was we
22:13
was we
22:13
was we said we're going to take this one
22:16
said we're going to take this one
22:16
said we're going to take this one service and we're going to break it
22:18
service and we're going to break it
22:18
service and we're going to break it into 15 smaller services so that anyone
22:22
into 15 smaller services so that anyone
22:22
into 15 smaller services so that anyone else in the company this is a large
22:24
else in the company this is a large
22:24
else in the company this is a large company of over a billion a year in
22:27
company of over a billion a year in
22:27
company of over a billion a year in Revenue um
22:30
so everyone can use every little piece
22:33
so everyone can use every little piece
22:33
so everyone can use every little piece so if anyone wants book data they've got
22:35
so if anyone wants book data they've got
22:35
so if anyone wants book data they've got it if anyone wants this they've got it
22:37
it if anyone wants this they've got it
22:37
it if anyone wants this they've got it if anyone wants this so we designed with
22:40
if anyone wants this so we designed with
22:40
if anyone wants this so we designed with that in mind and the reality was no one
22:45
that in mind and the reality was no one
22:45
that in mind and the reality was no one in the company used any of the
22:48
in the company used any of the
22:48
in the company used any of the microservices and it was because the
22:50
microservices and it was because the
22:50
microservices and it was because the corporate culture was not that anyone
22:52
corporate culture was not that anyone
22:52
corporate culture was not that anyone would use a microservice like we weren't
22:54
would use a microservice like we weren't
22:54
would use a microservice like we weren't at we weren't Amazon back in the early
22:57
at we weren't Amazon back in the early
22:57
at we weren't Amazon back in the early 2000s saying everything's a microservice
23:00
2000s saying everything's a microservice
23:00
2000s saying everything's a microservice you have to do this it was us in a
23:03
you have to do this it was us in a
23:03
you have to do this it was us in a little project thinking too big and
23:06
little project thinking too big and
23:06
little project thinking too big and again trying to boil the ocean
23:09
again trying to boil the ocean
23:09
again trying to boil the ocean um and when you talk about
23:11
um and when you talk about
23:11
um and when you talk about maintainability each service had its own
23:14
maintainability each service had its own
23:14
maintainability each service had its own repository so we now had 19 total
23:18
repository so we now had 19 total
23:18
repository so we now had 19 total repositories it it was confusing um now
23:22
repositories it it was confusing um now
23:22
repositories it it was confusing um now the code itself is is kind of nice um
23:26
the code itself is is kind of nice um
23:26
the code itself is is kind of nice um but we tried to do too much and it it
23:29
but we tried to do too much and it it
23:29
but we tried to do too much and it it cost us we probably spent we could have
23:31
cost us we probably spent we could have
23:31
cost us we probably spent we could have saved two to three months had we just
23:33
saved two to three months had we just
23:33
saved two to three months had we just not done
23:36
not done
23:36
not done that um so I talked about this you know
23:39
that um so I talked about this you know
23:39
that um so I talked about this you know microservices is a large scale
23:41
microservices is a large scale
23:41
microservices is a large scale organization decision if you're told to
23:43
organization decision if you're told to
23:43
organization decision if you're told to do something in microservices just ask
23:45
do something in microservices just ask
23:45
do something in microservices just ask like why like does this need to be a
23:48
like why like does this need to be a
23:48
like why like does this need to be a microservice microservices are great and
23:50
microservice microservices are great and
23:50
microservice microservices are great and if someone's going to use them as such
23:52
if someone's going to use them as such
23:52
if someone's going to use them as such awesome but otherwise decoupling all
23:54
awesome but otherwise decoupling all
23:54
awesome but otherwise decoupling all your code so you have to have projects
23:57
your code so you have to have projects
23:57
your code so you have to have projects that reference you have to have maybe a
23:58
that reference you have to have maybe a
23:58
that reference you have to have maybe a nougat package with all your object
24:00
nougat package with all your object
24:01
nougat package with all your object models and it's just it's excessive if
24:04
models and it's just it's excessive if
24:04
models and it's just it's excessive if you don't need
24:05
you don't need
24:06
you don't need it um we did test driven development
24:08
it um we did test driven development
24:08
it um we did test driven development that worked really well I wish we had
24:10
that worked really well I wish we had
24:10
that worked really well I wish we had tested it out a little bit more um in
24:12
tested it out a little bit more um in
24:12
tested it out a little bit more um in the front uh the other thing that was
24:15
the front uh the other thing that was
24:15
the front uh the other thing that was great to learn because I've been doing
24:17
great to learn because I've been doing
24:17
great to learn because I've been doing this for 25 years is high quality code
24:19
this for 25 years is high quality code
24:19
this for 25 years is high quality code is almost infinitely scalable so I come
24:23
is almost infinitely scalable so I come
24:23
is almost infinitely scalable so I come from the days where you can have a
24:25
from the days where you can have a
24:25
from the days where you can have a script that like a single server can
24:28
script that like a single server can
24:28
script that like a single server can only run a few hundred copies of it at a
24:30
only run a few hundred copies of it at a
24:30
only run a few hundred copies of it at a time you know so you need to have
24:32
time you know so you need to have
24:32
time you know so you need to have replication and so like there like every
24:35
replication and so like there like every
24:35
replication and so like there like every CPU cycle
24:37
CPU cycle
24:37
CPU cycle mattered in today's world of cloud
24:41
mattered in today's world of cloud
24:41
mattered in today's world of cloud computing it it it almost doesn't matter
24:43
computing it it it almost doesn't matter
24:43
computing it it it almost doesn't matter like we had we set up this
24:48
like we had we set up this
24:48
like we had we set up this microservices to be able to scale up to
24:51
microservices to be able to scale up to
24:51
microservices to be able to scale up to 50
24:52
50
24:52
50 instances and we would have two was our
24:55
instances and we would have two was our
24:55
instances and we would have two was our maximum we ever hit and we were Sur
24:58
maximum we ever hit and we were Sur
24:58
maximum we ever hit and we were Sur reporting two to 300,000 users a
25:01
reporting two to 300,000 users a
25:01
reporting two to 300,000 users a day so it was like we just it was NE
25:04
day so it was like we just it was NE
25:04
day so it was like we just it was NE performance was never an
25:06
performance was never an
25:06
performance was never an issue um also if you work for a large or
25:09
issue um also if you work for a large or
25:09
issue um also if you work for a large or medium-sized company security and
25:11
medium-sized company security and
25:11
medium-sized company security and compliance needs to be part of your
25:13
compliance needs to be part of your
25:13
compliance needs to be part of your requirements and planning um the amount
25:15
requirements and planning um the amount
25:15
requirements and planning um the amount of time we spent getting three quarters
25:17
of time we spent getting three quarters
25:17
of time we spent getting three quarters the way through something and legal
25:19
the way through something and legal
25:19
the way through something and legal would come in and tell us it had to work
25:22
would come in and tell us it had to work
25:22
would come in and tell us it had to work differently again plan plan
25:25
differently again plan plan
25:25
differently again plan plan plan all right for every one hour spent
25:28
plan all right for every one hour spent
25:28
plan all right for every one hour spent planning you save 10 hours of
25:30
planning you save 10 hours of
25:30
planning you save 10 hours of development um I started out as a
25:33
development um I started out as a
25:33
development um I started out as a developer and I'm going to speak for
25:36
developer and I'm going to speak for
25:36
developer and I'm going to speak for myself but I know I'm not alone I love
25:38
myself but I know I'm not alone I love
25:39
myself but I know I'm not alone I love to kind of dive in and code I love to
25:42
to kind of dive in and code I love to
25:42
to kind of dive in and code I love to see success early I love to see things
25:46
see success early I love to see things
25:46
see success early I love to see things move and planning is hard for me like I
25:49
move and planning is hard for me like I
25:49
move and planning is hard for me like I have to tell myself stop and plan but
25:53
have to tell myself stop and plan but
25:53
have to tell myself stop and plan but that said and I spend more time planning
25:57
that said and I spend more time planning
25:57
that said and I spend more time planning when you think you're done planning plan
25:58
when you think you're done planning plan
25:58
when you think you're done planning plan some
25:59
some
25:59
some work and it
26:04
work and it
26:04
work and it organizationally it can be hard because
26:06
organizationally it can be hard because
26:06
organizationally it can be hard because it's like you've spent two weeks what
26:07
it's like you've spent two weeks what
26:07
it's like you've spent two weeks what have you done and it's like I wrote two
26:09
have you done and it's like I wrote two
26:09
have you done and it's like I wrote two pages in a planning document it doesn't
26:11
pages in a planning document it doesn't
26:11
pages in a planning document it doesn't sound like a lot um but if it's like
26:14
sound like a lot um but if it's like
26:14
sound like a lot um but if it's like I've evaluated four different Frameworks
26:17
I've evaluated four different Frameworks
26:17
I've evaluated four different Frameworks I've you know thought about our data
26:18
I've you know thought about our data
26:18
I've you know thought about our data model and tweaked it 1,700 times because
26:21
model and tweaked it 1,700 times because
26:21
model and tweaked it 1,700 times because I found this one thing and like those
26:24
I found this one thing and like those
26:24
I found this one thing and like those things are going to save you months of
26:27
things are going to save you months of
26:27
things are going to save you months of effort so so plan plan
26:31
effort so so plan plan
26:31
effort so so plan plan plan um I have one more project that I
26:34
plan um I have one more project that I
26:34
plan um I have one more project that I kind of want to talk a little bit about
26:35
kind of want to talk a little bit about
26:35
kind of want to talk a little bit about but I think I'll come back to that if I
26:36
but I think I'll come back to that if I
26:36
but I think I'll come back to that if I have
26:37
have
26:37
have time um 90% of apps are over engineered
26:40
time um 90% of apps are over engineered
26:41
time um 90% of apps are over engineered so I'm gonna say this you know again
26:43
so I'm gonna say this you know again
26:43
so I'm gonna say this you know again this is not a duplicate showing again
26:44
this is not a duplicate showing again
26:44
this is not a duplicate showing again for effort
26:47
for effort
26:47
for effort um I have not seen enough planning and
26:51
um I have not seen enough planning and
26:51
um I have not seen enough planning and especially the planning where the
26:52
especially the planning where the
26:52
especially the planning where the business and the tech side are really
26:54
business and the tech side are really
26:54
business and the tech side are really engaging early on um
26:58
engaging early on um
26:58
engaging early on um it you'll save so much time down the
27:02
it you'll save so much time down the
27:02
it you'll save so much time down the road by
27:05
road by
27:05
road by planning and not over
27:08
planning and not over
27:09
planning and not over engineering all right and again you are
27:11
engineering all right and again you are
27:11
engineering all right and again you are paid to solve problems not to write code
27:14
paid to solve problems not to write code
27:14
paid to solve problems not to write code um if you are asked to build an
27:18
um if you are asked to build an
27:18
um if you are asked to build an e-commerce
27:20
e-commerce
27:20
e-commerce website because you want to sell
27:22
website because you want to sell
27:22
website because you want to sell handmade
27:24
handmade
27:24
handmade crafts if you're not using Shopify or
27:27
crafts if you're not using Shopify or
27:27
crafts if you're not using Shopify or Etsy you're making a bad financial
27:31
Etsy you're making a bad financial
27:31
Etsy you're making a bad financial decision with a little asterisk there
27:34
decision with a little asterisk there
27:34
decision with a little asterisk there yes if you have some crazy new technique
27:36
yes if you have some crazy new technique
27:36
yes if you have some crazy new technique or something else going on fine the cost
27:40
or something else going on fine the cost
27:40
or something else going on fine the cost that Etsy or Shopify is going to take
27:43
that Etsy or Shopify is going to take
27:43
that Etsy or Shopify is going to take from your
27:45
from your
27:45
from your transaction is not anywhere close to the
27:48
transaction is not anywhere close to the
27:48
transaction is not anywhere close to the cost of developing maintaining and
27:50
cost of developing maintaining and
27:50
cost of developing maintaining and dealing with all the other hassles so as
27:53
dealing with all the other hassles so as
27:53
dealing with all the other hassles so as a technologist as a technology manager
27:57
a technologist as a technology manager
27:57
a technologist as a technology manager it's your job to what's the problem
27:59
it's your job to what's the problem
27:59
it's your job to what's the problem you're trying to
28:00
you're trying to
28:00
you're trying to solve and how can you solve it not
28:03
solve and how can you solve it not
28:03
solve and how can you solve it not necessarily how can you solve it with
28:08
code all right
28:10
code all right
28:10
code all right so design is for tomorrow and not 10
28:14
so design is for tomorrow and not 10
28:14
so design is for tomorrow and not 10 years from now especially with how fast
28:16
years from now especially with how fast
28:16
years from now especially with how fast AI is moving with how fast cloud
28:18
AI is moving with how fast cloud
28:19
AI is moving with how fast cloud computing is moving
28:22
computing is moving
28:22
computing is moving um
28:24
um
28:24
um you if you're designing a solution
28:27
you if you're designing a solution
28:27
you if you're designing a solution that's going to work work for 10 years
28:30
that's going to work work for 10 years
28:30
that's going to work work for 10 years it's probably not um and one of the
28:33
it's probably not um and one of the
28:33
it's probably not um and one of the previous uh one of the previous
28:35
previous uh one of the previous
28:35
previous uh one of the previous presentations showed a wonderful quote
28:38
presentations showed a wonderful quote
28:38
presentations showed a wonderful quote um about you know being fluid uh and the
28:41
um about you know being fluid uh and the
28:41
um about you know being fluid uh and the reality is technology is changing a lot
28:44
reality is technology is changing a lot
28:44
reality is technology is changing a lot and the best thing you can be is Nimble
28:47
and the best thing you can be is Nimble
28:47
and the best thing you can be is Nimble and small and fluid so that you can
28:49
and small and fluid so that you can
28:49
and small and fluid so that you can adapt to whatever is coming up so the
28:51
adapt to whatever is coming up so the
28:51
adapt to whatever is coming up so the build versus buy if you have
28:55
build versus buy if you have
28:55
build versus buy if you have to buy a new piece of software every
28:58
to buy a new piece of software every
28:58
to buy a new piece of software every year it's probably still cheaper than
29:01
year it's probably still cheaper than
29:01
year it's probably still cheaper than hiring a developer to maintain an update
29:03
hiring a developer to maintain an update
29:03
hiring a developer to maintain an update piece of
29:05
piece of
29:05
piece of software um the the Middle Point um has
29:10
software um the the Middle Point um has
29:10
software um the the Middle Point um has been called out to me is really
29:11
been called out to me is really
29:11
been called out to me is really important developers are expensive
29:14
important developers are expensive
29:14
important developers are expensive infrastructure is
29:15
infrastructure is
29:15
infrastructure is cheap so what that kind of means is
29:21
cheap so what that kind of means is
29:21
cheap so what that kind of means is that if you're so this goes back to
29:24
that if you're so this goes back to
29:24
that if you're so this goes back to build versus buy if there's a solution
29:26
build versus buy if there's a solution
29:26
build versus buy if there's a solution out there that costs
29:30
$250 that's you know some in many cases
29:35
$250 that's you know some in many cases
29:35
$250 that's you know some in many cases less than a developers day worth of
29:37
less than a developers day worth of
29:37
less than a developers day worth of effort chances are a $250 piece of
29:40
effort chances are a $250 piece of
29:40
effort chances are a $250 piece of software is going to take a lot more
29:42
software is going to take a lot more
29:42
software is going to take a lot more than a day to to manage or
29:44
than a day to to manage or
29:44
than a day to to manage or maintain so you know where possible you
29:49
maintain so you know where possible you
29:49
maintain so you know where possible you know buy things that are
29:50
know buy things that are
29:50
know buy things that are pre-made also with
29:53
pre-made also with
29:53
pre-made also with infrastructure optimizing a database so
29:56
infrastructure optimizing a database so
29:56
infrastructure optimizing a database so that you only need one instance of a
29:58
that you only need one instance of a
29:58
that you only need one instance of a database versus two because you don't
29:59
database versus two because you don't
29:59
database versus two because you don't want replication or things like that you
30:02
want replication or things like that you
30:02
want replication or things like that you can replicate a database for $250 you
30:05
can replicate a database for $250 you
30:05
can replicate a database for $250 you know $500 a month of cloud a developer
30:08
know $500 a month of cloud a developer
30:08
know $500 a month of cloud a developer costs a lot more than that so spending a
30:10
costs a lot more than that so spending a
30:10
costs a lot more than that so spending a month trying to optimize code is usually
30:14
month trying to optimize code is usually
30:14
month trying to optimize code is usually not worth it when you can just throw
30:15
not worth it when you can just throw
30:15
not worth it when you can just throw Hardware at it um also easy to read
30:20
Hardware at it um also easy to read
30:20
Hardware at it um also easy to read code um over performance because a
30:24
code um over performance because a
30:24
code um over performance because a senior developer who writes code that
30:26
senior developer who writes code that
30:26
senior developer who writes code that only a senior developer can edit is
30:30
only a senior developer can edit is
30:30
only a senior developer can edit is going to cost you time when that senior
30:32
going to cost you time when that senior
30:32
going to cost you time when that senior developer is gone and you have a mid or
30:34
developer is gone and you have a mid or
30:34
developer is gone and you have a mid or a junior spending a month trying to
30:36
a junior spending a month trying to
30:36
a junior spending a month trying to figure out what that code did um so
30:39
figure out what that code did um so
30:39
figure out what that code did um so that's where the simplification process
30:41
that's where the simplification process
30:41
that's where the simplification process comes in that's where planning comes in
30:45
comes in that's where planning comes in
30:45
comes in that's where planning comes in because the amount of times I've seen a
30:47
because the amount of times I've seen a
30:47
because the amount of times I've seen a project where they chose some esoteric
30:51
project where they chose some esoteric
30:51
project where they chose some esoteric package or esoteric framework that some
30:54
package or esoteric framework that some
30:55
package or esoteric framework that some senior design developer who probably
30:57
senior design developer who probably
30:57
senior design developer who probably contributed to that
30:59
contributed to that
30:59
contributed to that framework did and chose and it may be
31:02
framework did and chose and it may be
31:02
framework did and chose and it may be elegant it may be quick it may be super
31:04
elegant it may be quick it may be super
31:04
elegant it may be quick it may be super performant but you can't get other
31:06
performant but you can't get other
31:06
performant but you can't get other developers to work in that environment
31:08
developers to work in that environment
31:09
developers to work in that environment so you're it's short-term decisions that
31:13
so you're it's short-term decisions that
31:13
so you're it's short-term decisions that hurt the
31:14
hurt the
31:14
hurt the longterm um the last Point here is
31:16
longterm um the last Point here is
31:17
longterm um the last Point here is momentum is hard to break if you get a
31:19
momentum is hard to break if you get a
31:19
momentum is hard to break if you get a developer you know group of developers
31:21
developer you know group of developers
31:21
developer you know group of developers Who start working in react they're not
31:24
Who start working in react they're not
31:24
Who start working in react they're not going to want to change they're not
31:26
going to want to change they're not
31:26
going to want to change they're not going to want to go to angular
31:28
going to want to go to angular
31:28
going to want to go to angular um so once you kind of commit to
31:30
um so once you kind of commit to
31:30
um so once you kind of commit to something it's very hard to get people
31:33
something it's very hard to get people
31:33
something it's very hard to get people that want to redo the same work they've
31:35
that want to redo the same work they've
31:35
that want to redo the same work they've done in something else so that's where
31:38
done in something else so that's where
31:38
done in something else so that's where in that planning and requirements it's
31:40
in that planning and requirements it's
31:41
in that planning and requirements it's really important to give your developers
31:43
really important to give your developers
31:43
really important to give your developers like a set deadline hey you've got one
31:45
like a set deadline hey you've got one
31:45
like a set deadline hey you've got one week install view install react install
31:48
week install view install react install
31:48
week install view install react install angular and then you tell me which one
31:51
angular and then you tell me which one
31:51
angular and then you tell me which one you think is
31:53
you think is
31:53
you think is best and then you evaluate and make your
31:56
best and then you evaluate and make your
31:56
best and then you evaluate and make your decision things like that
32:00
bad decisions are made with good
32:03
bad decisions are made with good
32:03
bad decisions are made with good intentions uh no one on the business
32:05
intentions uh no one on the business
32:05
intentions uh no one on the business side wants to hurt
32:08
side wants to hurt
32:08
side wants to hurt developers so when they make a request
32:11
developers so when they make a request
32:11
developers so when they make a request they're not doing it thinking oh this is
32:12
they're not doing it thinking oh this is
32:12
they're not doing it thinking oh this is going to be hard for the developers or
32:14
going to be hard for the developers or
32:14
going to be hard for the developers or this is going to make someone's life
32:15
this is going to make someone's life
32:15
this is going to make someone's life difficult they're just thinking this is
32:17
difficult they're just thinking this is
32:17
difficult they're just thinking this is what we
32:18
what we
32:18
what we want um you know and often times again
32:22
want um you know and often times again
32:22
want um you know and often times again choosing like a front end uh framework
32:24
choosing like a front end uh framework
32:24
choosing like a front end uh framework you'll decide well let's go with react
32:26
you'll decide well let's go with react
32:26
you'll decide well let's go with react because I have two react developers
32:28
because I have two react developers
32:28
because I have two react developers right now if you need SEO you need to go
32:32
right now if you need SEO you need to go
32:32
right now if you need SEO you need to go with next which yes is sort of react but
32:36
with next which yes is sort of react but
32:36
with next which yes is sort of react but you need to make sure that you're making
32:40
you need to make sure that you're making
32:40
you need to make sure that you're making the
32:41
the
32:41
the decisions
32:44
decisions
32:44
decisions correctly even though you might have
32:46
correctly even though you might have
32:46
correctly even though you might have good intentions you still might make bad
32:49
good intentions you still might make bad
32:49
good intentions you still might make bad decisions all right so let's go back to
32:51
decisions all right so let's go back to
32:51
decisions all right so let's go back to the planning phase because this is
32:53
the planning phase because this is
32:53
the planning phase because this is really where I think you this is what I
32:55
really where I think you this is what I
32:55
really where I think you this is what I think is the meat and potatoes so
32:57
think is the meat and potatoes so
32:57
think is the meat and potatoes so understand the requirements um obviously
33:01
understand the requirements um obviously
33:01
understand the requirements um obviously the earlier you can be part of that
33:03
the earlier you can be part of that
33:03
the earlier you can be part of that process the better but uh really what it
33:06
process the better but uh really what it
33:06
process the better but uh really what it comes down to is making sure that every
33:09
comes down to is making sure that every
33:09
comes down to is making sure that every developer who's writing code understands
33:12
developer who's writing code understands
33:12
developer who's writing code understands why they're doing what they're doing and
33:13
why they're doing what they're doing and
33:13
why they're doing what they're doing and what the requirements
33:15
what the requirements
33:16
what the requirements really evaluate many code framework
33:18
really evaluate many code framework
33:18
really evaluate many code framework options if you even need a framework um
33:22
options if you even need a framework um
33:22
options if you even need a framework um don't stop at the first one and uh you
33:26
don't stop at the first one and uh you
33:26
don't stop at the first one and uh you know the first good answer is not always
33:28
know the first good answer is not always
33:28
know the first good answer is not always the best answer have dialogue with your
33:31
the best answer have dialogue with your
33:31
the best answer have dialogue with your stakeholders constantly be talking um
33:34
stakeholders constantly be talking um
33:34
stakeholders constantly be talking um you know if you have a big de
33:36
you know if you have a big de
33:36
you know if you have a big de development team let's say you have 15
33:38
development team let's say you have 15
33:38
development team let's say you have 15 developers one or two Dev managers and
33:40
developers one or two Dev managers and
33:40
developers one or two Dev managers and then a director make sure that the
33:42
then a director make sure that the
33:42
then a director make sure that the director understands everything and the
33:44
director understands everything and the
33:44
director understands everything and the director then communicates to the
33:46
director then communicates to the
33:46
director then communicates to the manager make sure the managers
33:48
manager make sure the managers
33:48
manager make sure the managers understand things and the managers can
33:50
understand things and the managers can
33:50
understand things and the managers can you know make sure the uh developers
33:53
you know make sure the uh developers
33:53
you know make sure the uh developers understand things but everyone has to
33:55
understand things but everyone has to
33:55
understand things but everyone has to understand that
33:58
understand that
33:58
understand that um again the best option is not always
34:00
um again the best option is not always
34:00
um again the best option is not always the first
34:01
the first
34:01
the first one good option is good best option is
34:05
one good option is good best option is
34:05
one good option is good best option is best and research out of the box
34:07
best and research out of the box
34:07
best and research out of the box Solutions again you are not paid to
34:09
Solutions again you are not paid to
34:09
Solutions again you are not paid to write code you are paid to solve
34:13
problems uh understand the current
34:16
problems uh understand the current
34:16
problems uh understand the current future and wish list um I will also
34:18
future and wish list um I will also
34:19
future and wish list um I will also differentiate here where I did it before
34:21
differentiate here where I did it before
34:21
differentiate here where I did it before so the current requirements is what you
34:22
so the current requirements is what you
34:22
so the current requirements is what you need
34:23
need
34:23
need today um so if you were developing
34:26
today um so if you were developing
34:26
today um so if you were developing PayPal it's like the current
34:28
PayPal it's like the current
34:28
PayPal it's like the current requirements is people need to be able
34:29
requirements is people need to be able
34:29
requirements is people need to be able to transfer money to other people great
34:32
to transfer money to other people great
34:32
to transfer money to other people great future requirement might be we need to
34:34
future requirement might be we need to
34:34
future requirement might be we need to support contact list tra you know
34:36
support contact list tra you know
34:36
support contact list tra you know payments so we need to you do all the
34:38
payments so we need to you do all the
34:38
payments so we need to you do all the Taps a wish list requirement might be we
34:42
Taps a wish list requirement might be we
34:42
Taps a wish list requirement might be we want biometric
34:44
want biometric
34:44
want biometric protections so understanding that's it's
34:47
protections so understanding that's it's
34:47
protections so understanding that's it's like okay that's a big one we need to
34:49
like okay that's a big one we need to
34:49
like okay that's a big one we need to keep an eye out and we need to be aware
34:52
keep an eye out and we need to be aware
34:52
keep an eye out and we need to be aware of it but it's not something we need to
34:53
of it but it's not something we need to
34:54
of it but it's not something we need to design a whole system to be able to work
34:56
design a whole system to be able to work
34:56
design a whole system to be able to work with it um because it might be too big
34:59
with it um because it might be too big
34:59
with it um because it might be too big it might be out of scope we might not
35:01
it might be out of scope we might not
35:01
it might be out of scope we might not have the technology available today to
35:03
have the technology available today to
35:03
have the technology available today to be able to do it in a reasonable
35:07
way all right so the key takeaways uh
35:10
way all right so the key takeaways uh
35:10
way all right so the key takeaways uh repeat repeat repeat you are paid to
35:13
repeat repeat repeat you are paid to
35:13
repeat repeat repeat you are paid to solve problems not to write code um for
35:16
solve problems not to write code um for
35:16
solve problems not to write code um for every one hour you spend planning you
35:18
every one hour you spend planning you
35:18
every one hour you spend planning you save 10 hours of execution and less is
35:22
save 10 hours of execution and less is
35:22
save 10 hours of execution and less is more so unless you have the weird job
35:25
more so unless you have the weird job
35:25
more so unless you have the weird job where you're paid per line of code
35:28
where you're paid per line of code
35:28
where you're paid per line of code um writing less code that solves the
35:31
um writing less code that solves the
35:31
um writing less code that solves the problem still is is going to be more
35:34
problem still is is going to be more
35:34
problem still is is going to be more valuable to yourself and the other
35:36
valuable to yourself and the other
35:36
valuable to yourself and the other people on your team um and you should
35:39
people on your team um and you should
35:39
people on your team um and you should work towards
35:41
that all right so thank you I think we
35:44
that all right so thank you I think we
35:44
that all right so thank you I think we have some time for questions if not I
35:46
have some time for questions if not I
35:46
have some time for questions if not I will go back to my uh other story
35:53
[Music]

KISS - Keep It Simple Stupid | Software Architecture Conference

c-sharpcorner_com

What is New in GPT 5

Top 10 Vibe Coding Tools to Try Now

The Cloud Show with Magnus Mårtensson ft. Mahesh Chand - Ep: 73

Hacking MySQL for Big Data | Software Architecture Conference

Rockin’ The Code World with dotNetDave ft. Leslie Richardson - Ep. 94

Rockin’ The Code World with dotNetDave ft. Milan Jovanovic Ep. 76

Rockin’ The Code World with dotNetDave ft. Brady Gaster Ep. 75

Rockin' The Code World with dotNetDave ft. Mahesh Chand Ep 15

Boosting Productivity with .NET Source Generators

Welcome to FreeMediaTools – Build Web Apps from Scratch Coding & Programming Channel Intro

Build a Next.js & React.js PDF Editor in Browser Using jsPDF,Html2Canvas & Fabric.js Using JS

Build a React.js PDF Viewer & Editor With Controls Using pdfjs-dist Library in Browser Using JSX

Up next in 10

KISS - Keep It Simple Stupid | Software Architecture Conference

c-sharpcorner_com