Ruby Video | SF Bay Area Ruby Meetup

00:00:01.800 all right this is pretty awesome to see um yeah we're going to be starting uh we're going to be starting this

00:00:08.400 but I I want to say it's just so amazing to see people to see people chatting and

00:00:15.599 yeah not willing to stop and just connecting this is pretty awesome yeah

00:00:20.960 thank you all um thank you all for coming I I should speak up got it

00:00:30.400 you remember my rails c talk I can do should I do this

00:00:38.280 no you can control yeah all right awesome um so yeah uh we we're beginning

00:00:45.360 and um so first of all thank you all for coming my name is Serena uh huge shout

00:00:51.920 out to chim hosting us today

00:01:03.519 so cool very kind of non-standard space yeah and but most of all thank you

00:01:11.080 thank thank you to all of you for coming for connecting for being here and um

00:01:17.360 just kind of building this excitement for Ruby and rails and this

00:01:23.560 community uh that I'm enjoying so much you guys have no idea what it means uh

00:01:30.680 because since I moved to San Francisco in January this event has been the

00:01:36.399 highlight of my experience here and I just love uh seeing all of you and

00:01:42.799 seeing so many uh now already familiar faces and friends so thank you thank you

00:01:48.719 for coming so we're gonna have a pretty crazy uh agenda today because we have

00:01:55.479 Noel rapping whom whose books I'm sure you all know uh talking about um Rubik system and and

00:02:05.079 Noel works for chime so he will open with some intro about Chim

00:02:12.680 uh then we have my colleague Vadim de also a PO popular book author uh talking

00:02:20.160 about some beautiful patterns and then after the Break um we have an open mic

00:02:29.040 as always so prepare your short announcements

00:02:35.440 shoutouts if you're hiring if you're looking to get hired if you are just

00:02:41.280 discovered some gem or if you have a question please prepare those little

00:02:47.360 bits to share and finally we'll have a presentation from brandan which will be

00:02:54.200 a u like a sneak peek into rubikon keynote this year here so it's going to

00:03:01.280 be pretty amazing and uh finally after um after this uh we can go whomever can

00:03:11.959 we can go grab drinks in a place nearby so that's kind of an addition to the

00:03:18.040 agenda yeah I I just loved people did it last time and it was kind of cool after

00:03:23.920 GitHub so yeah with that uh let me just um uh

00:03:33.200 what happened what happened uh remind you

00:03:53.959 yeah okay this pause oh this pause is sponsored by evil Martians just just for

00:04:00.439 you to yeah just to watch our logos more so yeah I run evil martians we are on a

00:04:07.079 mission to support startups on Rails so we built open source I hope you've

00:04:12.319 recognize and use uh some of that we build uh we write content but also we support companies so if you have if

00:04:18.359 you're a real startup and you're struggling with something please reach out we'll fix those problems for you and

00:04:24.680 will help you be insanely successful just like ch

00:04:31.000 yeah so uh by the way final link um SF Ruby at Luma sign up this is where you

00:04:39.160 will see the uh next event um which I'm not disclosing the date yet uh but I'm

00:04:46.960 going to say we'll have a rockstar okay we'll have Jeremy Evans

00:04:52.280 right all right I said it yep um now no you're up

00:05:01.160 so we have this amazing uh book author

00:05:07.880 and um Ruby is Enthusiast I'm sure you uh you might have seen his presentations

00:05:13.880 at many conferences but today no traveled to San Francisco to meet all of

00:05:21.039 you and to what could be I I I have no idea like what could be better and more

00:05:27.880 amazing than to have no rapping speaking on behalf of Chim at our

00:05:36.800 Meetup I am thank you um thank you for that extremely generous introduction I am

00:05:42.680 going to try something ridiculous which is to run my presentation off of my phone for stupid Logistics reasons that

00:05:48.319 you don't care about um hopefully it will work if not I have at least two backup plans uh one of which is just

00:05:55.360 vamping anyway who needs slides slides are not important uh I want to give a shout out out to um the chime workspace

00:06:02.479 team that put together the food spread and set up the room and made sure all the AV and stuff was laid out uh just

00:06:08.880 really great people to work with um also we have a number of chime employees here

00:06:15.039 chime people please raise your hand um they will be happy to talk to you about

00:06:20.520 chime uh during the break or whenever uh you want to uh chime is a financial

00:06:26.680 services company uh our we do a ton of rubyan rails on our back end um we are

00:06:34.120 possibly hiring I honestly have no idea uh careers chime.com careers is a good

00:06:39.800 place to go we are hiring I'm sure we're hiring we're probably hiring

00:06:44.919 um and I uh if you are interested in making the financial World better for

00:06:51.360 people who could use uh a better Financial World um CH is a great place

00:06:56.599 to be uh so I'm sure all the time employees will be very happy happy to speak to you about that uh uh during the

00:07:02.840 break or afterwards so I wanted to thank them and uh you can talk among yourselves for like two seconds while I

00:07:08.720 figure out how to share my screen

00:07:23.000 here outstanding outstanding outstanding uh there we go

00:07:32.000 better there we go yeah all right very exciting uh the dumbest thing I've actually ever

00:07:39.360 done uh yeah normally I normally I leave this up while I introduce stuff uh so uh

00:07:46.120 but we'll just keep going okay so I had kind of a unique

00:07:53.440 experience with a legacy Ruby codebase uh a legacy Ruby project I guess it's

00:08:00.000 it's better to say um and that was of course this uh when I suddenly found myself contractually obligated to turn

00:08:07.840 the book on the left which is the fourth edition of the programming Ruby book which came out in 2013 into the book on

00:08:15.240 the right which is the uh fifth edition which came out last January this uh forced me uh so the so

00:08:22.720 sorry the existing version was a pretty good snapshot of the Ruby ecosystem as it existed in 2012 um much has changed

00:08:30.000 since then as I'm sure you can all understand uh in particular we had I think 11 versions of Ruby 2122 so on

00:08:38.120 through 3 three um and countless other uh versions of gems and tools and all

00:08:45.040 sorts of changes like that and that forced me to confront uh what had

00:08:51.120 changed in the ecosystem like had very pragmatic need to think both about what had changed and which of those changes

00:08:58.200 were most important uh because the goal of course is to change the book but not change the book

00:09:05.959 right we want it to still be the reference place for people beginning rubyists intermediate rubyist expert

00:09:12.399 rubius rubyists we still wanted to give that overview of the entire Ruby

00:09:17.959 ecosystem so um this is a metaphor I guess for legacy code um but it's not

00:09:24.360 just a metaphor um as you may know pragmatic press

00:09:30.079 uh has a build system for their books the books exist as markdown text used to be XML text a whole bunch of J Ruby code

00:09:37.519 actually that converts them to uh PDF and whatever the pickaxe book has some

00:09:42.800 special Ruby code all of the code samples in the pickax book are um

00:09:48.800 executed when you build the book so the output is directly inserted uh this is

00:09:54.480 as you might imagine kind of weird and esoteric Ruby code that's pulling random strings and throwing them to an

00:09:59.519 interpreter in messing with build pass this code was written by Dave Thomas probably in the year 2000 I don't know

00:10:05.920 for sure I never asked so like one of the smartest Ruby developers I have ever

00:10:10.959 met uh bequeathed me a bunch of code that was written 20 plus years ago that he never intended anybody else to see

00:10:18.680 and basically said I'll help you if something breaks so a legacy code project

00:10:25.959 literally and uh figuratively so we're going to talk about this this in terms of in terms of Ruby's Legacy and Ruby's

00:10:32.920 future as you know by now because it's been said multiple times I'm no rapping I work here at chime thank you all so

00:10:39.880 much for coming out I very much appreciate it I'm not done actually that's just the intro um so we'll go

00:10:45.680 from here so the perspective that I eventually came to was to think about Ruby's entire

00:10:55.160 ecosystem as a one big Legacy code base and the definition here of Legacy code

00:11:03.360 that I use is code where the context of the creation has been lost the Ruby ecosystem is thousands upon thousands of

00:11:10.200 decisions that were made under a particular set of constraints some of which are still applicable some of which

00:11:15.920 are not right the analogy I like to use is a parking lot how many times have you

00:11:21.200 gone into a parking lot seen somebody parked at like a 35 degree angle and you go what how could anybody be such bad at

00:11:27.480 parking that they would park that at that crazy angle and what actually happened is when they got there it was the only place left and the person next

00:11:34.160 to them was parked at a crazy angle that person is left and now all you have is

00:11:39.240 like the the one stupid looking car at the bad angle but you don't know what

00:11:44.440 the constraints were like when that person showed up and similarly we don't we can't we don't and can't know all of

00:11:51.600 the constraints that various people have put into various Ruby decisions over the years so the way I like to uh keep keep

00:11:59.480 that in mind so uh it is also like another feature of

00:12:06.440 Legacy code is it is risky to change because of hidden dependencies and if you think about the Ruby ecosystem we're

00:12:12.959 not even just talking about dependencies in terms of code breaking we're talking about dependencies that change the way

00:12:19.720 that Ruby developers think about Ruby if you've ever looked at a longstanding rails program you can be you can look at

00:12:26.800 it almost archaeologically in terms of at like that was the year we all cared about structs and uh that was the year

00:12:33.600 we used form objects uh and and you know all of these individual changes deeply

00:12:41.160 affect the way people interact with Ruby and with the ecosystem so I want to talk about like

00:12:48.320 specifically what changed and I'm sort of working towards a theory of what

00:12:54.320 changes take and what changes don't and I also just think it's interesting to

00:12:59.760 talk about what changed over the 12E period so the language itself changed

00:13:06.320 every version of Ruby has two three four syntax changes sometimes small sometimes

00:13:11.519 big I'm going to split those into like incremental changes and big changes

00:13:16.600 incremental changes we'll go through this real quickly this is a representative sample it is not exhaustive um there uh the uh Ruby

00:13:24.480 changes site and I am blanking on the person's name the uh who who runs it is uh absolutely outstanding at tracking

00:13:31.040 this stuff over time it was invaluable to me uh when putting this together so incremental changes in in Ruby 21 def

00:13:38.680 returns a symbol which offers this alternate way of basically allows decorator uh function decorators um we

00:13:46.839 had rational and complex literals 23 uh added the safe navigation operator a

00:13:52.680 common Ruby idiom on the top line that was made like easier to type um it also

00:13:58.600 added here do with indentation a less common Ruby idiom made significantly less annoying to type um 26 and 27 added

00:14:07.839 partial ranges in both directions you know these kinds of things 27 added numbered block progams endless method

00:14:14.680 syntax argument for writing the main point of this is that you get the idea here that there are a number of a number

00:14:20.759 of changes here along the lines of here is a common Ruby pattern that we are

00:14:26.279 going to make slightly easier to type there have also been big changes my

00:14:33.120 definition here of a big change was a change that was uh designed to really

00:14:38.440 change the way that you interact with Ruby as a programming language there was one that just missed refinements which

00:14:44.800 actually did make it into the previous book just barely um which was designed to change the way you do monkey patching

00:14:50.800 um there were in this time period uh there were three things that I classified as big changes uh pattern

00:14:57.720 matching um uh pattern matching is uh introduced in 27 augmented in 30 and Beyond is a how

00:15:05.680 many people actually use pattern matching on a regular basis raise your hands higher than that I can't see it if

00:15:11.360 you hide it behind the other person so that is a solid 5% of the

00:15:16.480 room uh what you want to you just got okay I want to I just

00:15:24.120 got on 27 good good good good luck um uh so pattern matching is an ambitious

00:15:30.600 attempt to change the way you deal with complicated data structures in Ruby it is not an incremental change it is a

00:15:36.600 dramatic change to allow you to massively simplify certain patterns of interaction uh I credit Brandon with one

00:15:43.800 of the first uh presentations I ever saw that had it make sense to me uh which was which was very helpful um so that's

00:15:51.480 one the second one is not quite in the same league the keyword argument separation I think of this as a big

00:15:57.120 change between 27 and 3 because because it was a breaking change if you've updated a large chunk of code from 27 to

00:16:04.199 30 you know that even though it wasn't really supposed to break that much code it actually did break quite a bit of

00:16:10.319 code um so that counts as a big change it it is designed to force you to think

00:16:16.040 about the separation between keyword and and positional arguments in Ruby and the

00:16:21.160 third one and I guess is RBS how many people here use RBS he said to No Hands

00:16:28.360 coing Up one hand in the back I would have I would have taken the under on zero

00:16:34.639 frankly uh uh oh well Brandon is like Brandon is sort of like sheepishly

00:16:42.199 casually no no determinedly raising a hand what okay that's a definite hand

00:16:47.639 there's two two two RBS users so a solid god um RBS just just to be

00:16:55.759 complete how many people here have ever used refinements in Ruby he tried how many people continue

00:17:03.440 to use refinements regularly it's a solid three uh I'm building this there's a reason why I'm

00:17:09.760 asking all of this uh at least I hope there is you you better hope there is I'm just GNA start asking random yes

00:17:18.039 or no questions anyway um RBS is an attempt to add uh static typing to Ruby

00:17:25.919 uh officially as opposed to something like sorbet um Matt somewhat famously does not like static typing so this is

00:17:32.919 kind of a way to have static typing without actually having it in the actual syntax

00:17:39.120 um and I think I don't know what else to say about this like there's been very little tooling and very little pickup of

00:17:44.679 it even as it has been out for two or three years at this point um so those

00:17:50.640 are the big those are the big language changes there have also been changes to the Ruby Library um every new version of

00:17:58.559 Ruby has a dozen or so small big changes to the Ruby standard Library often

00:18:03.840 individual methods I didn't bother to put all of these on separate slides but these are these are basically many of

00:18:10.200 these uh are mentioned in the book one of these actually obliterated an example in the book because it it made it a

00:18:16.559 oneliner instead of a cool use of how to use functional Pro how to chain functions um but you know these are this

00:18:24.240 is again just a representative sample every version of Ruby makes a bunch of changes like this they get they are

00:18:30.640 created typically for a very specific purpose and they are used for that specific purpose often you will pick one

00:18:36.960 of these things up and you won't have any idea how recently it's been added it's just like oh cool I can use match

00:18:43.360 question mark now I didn't know that was there uh there have been a couple of

00:18:48.679 changes that I would consider big changes from the Ruby Library again the same uh definition that library library

00:18:55.600 editions that are really meant to change the way that you interact act with Ruby and I I count

00:19:01.440 two uh rators how many people use

00:19:07.280 rators three thought about it once I used it for the book uh I learned them

00:19:14.640 for the book ra so here's the thing rators are an insanely complicated

00:19:19.679 technical a technical accomplishment in Ruby from the beginning of Ruby till the

00:19:24.880 introduction of rators one of the actual real constraints on Ruby programming was that it had a global interpreter lock

00:19:31.559 and only one piece of code could go through it at a time rators obliterated that like any time you take a

00:19:37.360 longstanding constraint and remove it there's going to be code that depends on it and and it's going to be subtle and

00:19:43.720 that happened here too it also invented an entirely new way to share code

00:19:48.960 between threads and Ruby that didn't exist before and only exists to share code within Raptors and for all that

00:19:56.039 work we have three people in this room who have used it um

00:20:03.080 uh they actually I mean they're not they're they we we'll we'll come back to this uh the other thing is not anywhere

00:20:10.240 near as big a change in the library but I think it's in its own way a very ambitious attempt to change the way that

00:20:16.320 you interact with Ruby and that is the data Library the data object how many people are using data objects I'm going

00:20:21.880 to bet that's a few more so that's about 10 or 12 uh so data um is an attempt to

00:20:28.960 give you an object that is as immutable as an object can be in Ruby um with and

00:20:34.919 although it is a fairly small set of code it has a very ambitious uh mandate

00:20:40.760 to really encourage people to use immutable value objects uh rather than

00:20:46.280 hashes and strings for their data so that that counts to me as a big a potentially big change okay um okay

00:20:54.200 category three I swear there's only five categories we're not going to be going through this forever um Category 3 is

00:20:59.600 implementation changes most of these didn't affect the book they don't really affect the ecosystem directly um it is

00:21:06.400 interesting that uh at least uh two of these four changes this is again a representative sample every version of

00:21:12.640 Ruby has minor minor performance and incr and and implementation changes these are the ones that seem most

00:21:17.760 important to me it's worth noting that both the symbol change and the yit change specifically have the goal of

00:21:23.679 making long running Ruby processes easier to manage and more performant uh as as kind of a trend in the Ruby

00:21:29.960 community over this time um I think prism it's not really clear what the impact of prism is going to be prism is

00:21:36.600 the new Ruby parser introduced in 33 it's going to be default in Ruby 34 um

00:21:42.000 it's not clear what the impact of that is going to be we'll talk a little bit more about that in a second but but I I

00:21:47.240 think it's going to be a hopefully a big change um the ecosystem around Ruby the

00:21:54.679 language has changed in a bajillion ways uh this is I think a complete list of

00:22:00.520 third-party tools that are mentioned in the pickaxe book in the new version of the pickaxe book uh bundler uh debugger

00:22:06.559 and pry rubocop and standard sorbet rack Sinatra WM yard it is worth noting that

00:22:12.440 the majority of these tools either did not exist in 2012 or were just getting started and had not achieved M share uh

00:22:19.360 in 2012 so bundler was mentioned in passing I think pry may have been mentioned in passing uh most of the

00:22:25.360 other ones did not exist um and many of these tools are like day-to-day pieces

00:22:31.960 of of people's life like how many people are using bundler it's going to be more than 10 hands right this is this is a

00:22:37.559 test of how many people are willing to raise their hands in public frankly uh you know how many people C I'm just

00:22:43.679 curious how many people are using sorbet on any kind of basis so that's 15 10 12 seven I don't know I

00:22:51.200 just keep saying numbers until one of them seems reasonable more than RBS is I think the relevant Point here um uh so

00:22:59.200 you know many of these changes really do affect the way that we work like bundler had a profound impact on the way Ruby

00:23:05.720 developers shared code um you know many of these have had other effects too there's obviously other third party

00:23:11.400 tools that you could mention that have had huge effects um and most importantly maybe

00:23:17.320 most importantly and sort of the hardest to quantify is how the Ruby Community itself has changed over that decade and

00:23:26.200 I this is I don't know I this is just more notes than anything else um a very strong change is the change in languages

00:23:34.120 that you that people are coming to Ruby from um especially the early versions of the pickaxe assumed that you knew Pearl

00:23:40.880 or Java uh nobody's coming to Ruby from Pearl anymore uh we are now getting

00:23:46.640 people mostly from Python and JavaScript and Java basing on my based on my own experience um that has a big impact on

00:23:55.279 how people write their Ruby code what idioms they're comfortable with what things seem strange to them it's a

00:24:00.679 different set of things um in this 10-year period we went through the whole boot camp uh rise and fall which had an

00:24:08.799 extremely profound impact on the the people the people who were coming into the Ruby Community especially in the

00:24:15.200 years when the boot camps were teaching people Ruby just a whole bunch of people came into the Ruby Community who might

00:24:20.919 not have been otherwise that have had like a tremendous and I think extremely positive impact on the community as a

00:24:26.399 whole um there's been a like emphasis on contributions from Big

00:24:32.440 players in the ecosystem especially Shopify but not just Shopify um and also

00:24:38.320 an emphasis on contributions that that um encourage long running rails program

00:24:45.080 kind of things um in a way that like the needs of a long running rails program are far different than the needs of a small scripting program uh and I think

00:24:52.840 recent changes in the Ruby ecosystem reflect that okay so where does that leave us um

00:24:59.600 a lot of those incremental changes have been adopted um as you can see like we have an extremely poor hit rate on those

00:25:06.399 big changes we had I listed five like big changes that came from the Ruby Community basically none of them had

00:25:13.760 majority usage in this group of Ruby enthusiasts um uh and I think that

00:25:21.120 that's that's interest and if you add ref like adding refinements that's six changes I guess the keyboard the keyword

00:25:27.559 thing everybody had to use so I guess that counts um but I think that's

00:25:32.720 interesting to me that that there have been all of these attempts to really uh

00:25:38.159 make dramatic changes to the Ruby ecosystem and they haven't to like a

00:25:43.440 first approximation have had very little effect whereas these small incremental

00:25:49.720 changes like I bet uh many of you use many of those small incremental changes

00:25:54.840 that I listed some probably more than others but I'm sure you found one two on that list that you use I'm not going to

00:26:00.440 go through them all with a show of hands that would be extremely tedious for everybody who is not me uh but I I do

00:26:06.080 think I think that there's something I think that there's something here in terms of like what kinds of changes are

00:26:11.960 adopted um but I also think it is interesting to think about like why

00:26:17.000 those big changes haven't been adopted and I think in several of those cases um

00:26:23.000 not data but definitely pattern matching and uh definitely RBS and definitely um

00:26:31.679 uh uh refinements and one of them that I'm just completely blanking on at the moment all were sort of introduced

00:26:38.640 Raptors sorry we're all introduced to the community in an explicitly not ready

00:26:43.679 for Primetime or experimental status which is great in terms of like getting

00:26:49.600 things out there and getting front of people at least in theory but I think meant that many people tried these

00:26:55.480 things or wanted to try these things and found that they were like not suitable

00:27:00.520 for what they were doing because they weren't finished yet or they didn't seem to be uh potentially they didn't seem to

00:27:07.399 be bottom up they seem to be imposed top down and sort of file them in the I don't need to know about this basket

00:27:13.880 which is a great basket to have trust me uh and never really came back to them um

00:27:20.360 it I think it's worth interesting it's it's worth contrasting a little bit with the rails ecosystem specifically the

00:27:26.159 rails ecosystem has reinvented itself a couple of times in the same time period and many of those changes to my like

00:27:32.080 informal bubble have been more widely adopted the rails team has is reinvented

00:27:37.760 their front-end implementation twice at least in that time both of those seem to have more ad option even the the really

00:27:44.799 new no build stimulus Hotwire stuff seems to have more pickup and play than any of those big Ruby changes um why I

00:27:53.480 think one of the reasons for that is that most of the time a big change in rail

00:27:58.640 comes only after it has been successful in a project where I think often big change in Ruby come from a more I don't

00:28:05.880 want to necessarily say academic perspective but from a definitely from a less of a battle tested perspective one

00:28:12.840 of the things that I am hopeful for in a world where the where prism is the Ruby parser is that it will enable more of a

00:28:19.640 playground mindset where you have the opportunity to turn on experimental

00:28:25.039 features or turn off experimental features and really exper experiment them without having them be official

00:28:31.640 parts of Ruby in a way that really isn't possible now um but I think might be really beneficial to the uh adoption of

00:28:39.679 these kind of new uh parts of Ruby as they go so uh all I've talked through

00:28:47.120 all of my slides I think incremental improvements have an up battle whatever new things survive when they solve

00:28:52.320 problems and I think one of the things that comes from the rails world is that a lot of their new things have already

00:28:58.200 solved a problem and they can say that and show those examples whereas a lot of the Ruby things are like these this

00:29:04.120 seems like a good idea and often they seem like really good ideas rators really do seem like a good idea uh as

00:29:09.840 somebody who is generally allergic to threading I find the metaphor of reactors like genuinely appealing and

00:29:15.679 relatively easy to get my head around and also I would never recommend it for a project at the moment um so because

00:29:23.840 it's just not quite there right so that's it so I I think that these ambitious projects within the Ruby

00:29:29.120 they're great but they need to they they need support they need support from us they need support from the Ruby Central

00:29:34.320 team hope the Ruby core team hopefully again we we'll have a world coming up where it will be easier to be more

00:29:40.840 experimental with these kinds of things um it's worth noting that Ruby 34 had a

00:29:45.880 syntax moratorium for the last six months or so uh the only major syntactic change in Ruby 34 is the introduction of

00:29:53.559 the keyword it to replace the underscore one uh which is minor but but also I'm

00:29:58.919 kind of looking forward to it um uh this was explicitly to allow the parsers

00:30:04.159 specifically prism to catch up um rather than having them try to hit a moving Target and we'll see where that goes we

00:30:10.600 also know from some things that Matt said I believe that this is true um that there will be a there's a major change

00:30:16.720 coming which may in fact turn out to be the breaking change that separates Ruby 3 from Ruby 4 which is the introduction

00:30:22.399 of Nam spaces being like modules the Nam separating out the namespace only feature of modules from the mod feature

00:30:28.399 of modules this may be coming there are a couple of different proposals I honestly don't pay a whole lot of attention to the Ruby proposals in draft

00:30:35.640 um but but uh this is one that that definitely came across my radar somebody else probably knows more somebody else

00:30:40.720 people somebody else here probably knows more about it but I think that that the question is how we can allow uh Ruby to

00:30:48.519 sustain continual change and be flexible and continue to evolve and grow you know

00:30:54.440 over the next 10 years as it has over the last 10 so try a new thing today we

00:31:00.320 I've given you many many options to choose from uh that's it that's all I've got I have no idea whether I was near

00:31:07.240 time on time under time uh I shouldn't turn my phone while it's sharing screen

00:31:12.320 and thank you for being a part of this this uh presentation experiment um programming Ruby is on sale uh you can

00:31:20.000 get it in ebook form from pragprog uh title ruby5 um there's an Amazon link

00:31:26.480 you can follow me on no rapid . um chime.com careers if you're interested

00:31:31.679 in chime in chime careers also separately if you are interested in writing for the pragmatic press please

00:31:39.480 talk to me I am also the Ruby series editor uh for pragmatic press and we're looking to get Ruby books in the

00:31:44.720 pipeline these do not have to be I promise they do not have to be 700 page bricks uh we're really they're really

00:31:50.320 leaning towards smaller uh books that you can turn around much faster uh if you have something that you are

00:31:56.120 passionate about in the Ruby community that you want to spend some time explaining to everybody in the Ruby Community uh please see me or contact me

00:32:03.720 somehow that is all I got uh I'm going to see if I can figure out how to turn off sharing here um but thank you very

00:32:10.240 much for your time thank you so much for coming out to our space we are so excited to host all of you here thank

00:32:15.639 you

00:32:29.679 wow that was amazing thank you thank you now right while vmer is very slowly

00:32:35.880 moving to the stage you're next

00:32:41.279 um yeah I'm just gonna do this I'm going to take a photo of you

00:32:48.760 all how about this okay this yep yep yep

00:32:58.880 all right um okay the next stop is

00:33:21.799 yeah um so always going to talk about something no mentioned

00:33:29.799 right you noticed what is

00:33:37.039 it you need

00:33:42.799 Power share and I can get you on the Wi-Fi yep go ahead wi know the password

00:33:53.519 y okay we'll take a minute

00:34:56.640 juste e

00:35:31.000 I think we're watching a movie

00:35:56.079 no yeah he wants to turn the lights off but no

00:36:02.480 no we're not doing it okay lights oh it it works you just need to

00:36:09.839 speak louder okay um okay

00:36:16.200 uh it's okay better okay hi everyone uh my name is Vladimir

00:36:24.400 and um want to share some thoughts on Rails and how we I don't

00:36:33.200 know try I would say try try to keep our code bases maintainable and some uh

00:36:39.440 particular use case actually I want to share today which deals with um

00:36:46.319 okay let me see um yeah um yeah the talk

00:36:52.640 today is forms on Rails and we're going to talk about I want to use use this as a particular

00:36:58.760 example of like modeling code and rails applications on a particular problem

00:37:04.319 dealing with complex forms and actually uh the use case I'm going to show we

00:37:09.960 just need to wait for the video to complete okay before I can switch to the next slide uh how

00:37:16.960 to fight complex UI while being a rails developer while

00:37:23.040 being a classic rails application developer not an API one with was front

00:37:28.400 end being a part of responsibility of another team uh but uh trying to be as

00:37:33.599 close to the classic rails to the HTML first approach and still being able to

00:37:39.720 solve challenges that sometimes we Face uh with with modern applications so this an an example

00:37:47.400 application in question let from the project called any cable plus it doesn't

00:37:53.440 matter what's important here is that that's how we kind of started like it's a rails startup so it's like rails

00:38:00.040 generate scaffold blah blah and and go on like and release and let users figure out what to do and we had some forms

00:38:07.760 there so this is we how we set up an any cable instance in the cloud and it

00:38:13.040 actually requires um decent amount of configuration settings it's not just the

00:38:19.040 whole format just a part of it and it contains like eight seven some inputs

00:38:24.960 and you need to scroll down you lose the context and it's not well the best user experience that's fine for rails

00:38:30.760 applications starting like you know just yesterday but you want to provide a

00:38:36.520 better user experience so we wanted to build something like this and that's what we've built in the end so we have a

00:38:43.640 form a multistep form or visard like we call it where you have multiple steps

00:38:49.640 you have some useful instructions on what you should add there at each step

00:38:55.240 and in the end you can submit it and create everything like a pretty common experience we you know uh learned back

00:39:03.839 in the windows days you know all windows installers looks like that so but how to

00:39:10.119 build this in rails so for our regular form we just created form for enumerate all the

00:39:17.960 fields like add some styling and that's it but now we have multiple steps back

00:39:24.119 and forth navigation we need to keep the state some how and so on we have a lot

00:39:29.560 of things to think about here and that's actually a lot of questions and I want to answer these questions uh by taking a

00:39:38.160 look at what is the Ra's way of doing things and how we

00:39:44.640 can kind of oh I would say enhance it to

00:39:49.839 make such non-trivial I would say forms uh possible but let's start with the rail

00:39:57.480 way as it looks from the form perspective so what is a form it's actually some request with some data and

00:40:04.200 this data goes through the layers of your rails abstractions so first we hit

00:40:09.680 the controller where we do something with the input filter data whatever then

00:40:15.000 we persist the data or invoke validations at the model level and then

00:40:20.319 we show some result using our view templates uh the question is where and

00:40:27.480 this configuration we must we should implement the multistep form the visard logic where it should live how we should

00:40:34.920 handle it well my answer would be is like well nowh we shouldn't try to fit

00:40:40.800 something that doesn't actually fit the classic rails Trio of MVC uh in we shouldn't fill it logic

00:40:47.839 into the classic MVC Trio we should try to add something else something in between your controller and model and

00:40:53.839 that's going to be something now already mentioned in his talk oh object they were popular someday and actually um

00:41:00.920 we're going to talk about the history a bit but the reason why we want to introduce something new and I think it's important to highlight here is this want

00:41:09.040 to separate the concerns the form and the logic we try to implement is actually UI driven logic it's not a part

00:41:15.520 of your model your model doesn't care how many steps do you need to fill the form it actually just needs the final

00:41:22.880 data user input to create something and invoke some business operations or

00:41:28.040 whatever and at the same time your controller also is not responsible for

00:41:33.760 thinking about this multi-step process it's only responsible for getting data

00:41:39.240 and passing it forward in a right format so we need something to separate these

00:41:44.520 concerns and without using rails concerns because they are well questionable um so we introduce firm

00:41:52.560 objects to deal with the following concerns the following responsibilities uh there are many of them they're just

00:41:59.400 the the most popular four it's about introducing context specific validations

00:42:04.680 because because not every validation should live at the model layer you only need to keep there something that's

00:42:11.000 required for your data consistency not for the particular use case sometimes we want to transform user

00:42:18.119 input because well your UI form might look not like your database table right

00:42:24.319 in most cases it is not we also need a way to provide a useful feedback to users that's an important role of this

00:42:30.760 intermediate abstraction it's not only about oneway uh data flow from user to

00:42:37.040 database it also about telling user what went wrong if something went wrong and

00:42:42.480 of course in our particular scenario we have some uid driven logic like

00:42:47.920 multistep form it also must have its own place in the code base but let's talk

00:42:53.119 about some past uh ideas with regards so

00:42:59.119 to adding something to rails to solve this problem and actually one of the

00:43:05.280 oldest mention of something that looked like form object comes from this gist by

00:43:10.880 dhh he called it aggregator but the idea is basically the same you have some form

00:43:17.280 in the interface that operates on multiple records or some like aggregate record like in this case we have a user

00:43:22.960 record and an email record let's create an agregator model whatever he wanted to

00:43:28.880 call it and use it in the controller like a regular model that's something

00:43:33.920 that looked like that never turned into life at least uh in rails or in uh

00:43:40.720 Source available 37 signal project but there was an attempt uh to actually

00:43:46.400 create an action form framework for rails uh about the same time like almost

00:43:52.640 seven eight years ago um well the attempt were considered

00:43:58.400 unsuccessful so that that's why we don't have it right in rails but it's an interesting project just to learn about

00:44:03.920 how PR prior prior Generations tried to solve this problem including the rails

00:44:09.839 score team uh some days uh like yeah we're doing archology right now so reform was

00:44:17.520 a popular tool to generate forms it's a part of a Trailblazer I would say package for

00:44:25.640 modeling business logic and ra application but the reform were usually used independently of other Trail based

00:44:32.400 features and it was quite popular but also didn't kind of survive till uh till

00:44:38.599 now but it was looked like well you have a form object you create and you

00:44:44.119 operated like a whatever regular model in Ruby uh finally last history bit here uh

00:44:52.800 the latest attempt to bring something uh that looked like a for object at least the foundation for

00:44:59.599 potential form object happened just a year ago like almost exactly a year ago

00:45:05.319 uh there was the idea of introducing active model base like in a active record base but for non database backed

00:45:11.680 models for like virtual models unfortunately uh the idea was rejected

00:45:19.040 and with the same reason as the action form was rejected railcare team is not quite happy yeah the railcare team

00:45:26.440 understand that there is the need for something like that there is a missing Point uh the missing um Link in

00:45:35.000 this like abstractions chain in rails for some use cases at least but still there is no good solution that would

00:45:41.839 satisfy the core team and probably particular members well uh let me share the form

00:45:51.160 object abstraction that I'm happy with and I'm using in my R project and

00:45:56.440 actually not just using but trying to promote it in this

00:46:02.240 book uh which has a chapter dedicated to the problem with a lot of examples and

00:46:08.960 some useful stuff you can check it out later I don't fortunately I don't have copies with me today maybe next

00:46:14.760 time so I'm not going to recall like the book it's doesn't make sense you can

00:46:20.559 download it from Torrance probably or just ask me well and we don't don't have

00:46:27.200 discounts now I don't know maybe we do I don't know I'm I'm bad on promoting stuff so

00:46:34.920 yeah I'm going to show some stuff so this is a form object for our original form was just like one screen all the

00:46:43.520 fields but here I how I would implement it using not the model but the form

00:46:49.720 object like uh using some application form we're going to talk about it in a minute uh let's take a closer look what

00:46:55.520 do we have here so so we have attributes that corresponds one to one to the

00:47:01.079 fields we have in the UI right and we have some Logic for initializing the

00:47:07.440 object we want to create and actually creating submitting a form so we use the

00:47:12.599 term submit because well that's what we do with forms well it doesn't bring a lot of

00:47:19.040 additional value to like uh compared to using a model the

00:47:25.079 only probably thing that we solve here is that we decouple UI fields from the

00:47:31.040 database schema so we can have different map like we have some specific mapping to map UI form values to the actual

00:47:38.880 model attributes and we don't need to update our UI or model to reflect one

00:47:44.720 another so that's better than nothing but still not enough to justify

00:47:49.760 introducing this abstraction right well in in reality we have more stuff here we

00:47:55.880 can add validations at this level without adding them to the model however

00:48:01.079 uh here have like validates name and region presence these two are probably

00:48:06.960 are invariance of the model they should be validated at the model L and we can

00:48:12.000 do this by just delegating validations to the model and that's fine that's how we avoid duplication with using when

00:48:18.520 using form objects we just uh add the validations that only make sense in this

00:48:23.680 particular context uh also we can introduce callbacks to trigger some side

00:48:30.960 operations or some other stuff that relates to this particular form compared to model callbacks this going to be

00:48:38.520 executed only in the context of this particular user action in this particular scenario so we do not need to

00:48:44.960 deal with the global callbacks defined in the model finally uh there is one

00:48:50.640 line that's hard to explain why do we have it what does it mean like self

00:48:55.720 model name cable and to talk about it we need to

00:49:02.119 learn the rules behind the abstraction the application form I use in

00:49:07.400 application and in the book and why I build it this way and the reasoning behind this is whatever you try to

00:49:14.960 introduce into the rails application though should play nicely with what

00:49:20.760 rails gives you you shouldn't play against the framework you should try to be a part of the team

00:49:27.640 that's actually some the first edition of the rail sway book uh also pretty old like almost 20 years old right now but

00:49:34.920 the phrase still holds what does it mean to play with

00:49:40.680 framework play the same team well I split this into several important points

00:49:47.720 first you need to know how rails Works actually under the hood why rails is designed this way which building blocks

00:49:54.920 rails relies on because there are many well not so many actually things and rails that's used throughout all the

00:50:02.160 components of the framework and you can easily find a pattern and try to use it in your code that that would make your

00:50:08.599 code look like the rails original code of course you must follow the

00:50:15.000 conventions conventions is the most powerful feature introduced by Rubi on

00:50:20.960 Rails and that's what you what makes you productive and productivity is a thing we uh care the most uh when building

00:50:30.359 with rails otherwise we wouldn't use rails right so that would make you code

00:50:35.400 play nicely with rails let's see uh how we achieve this uh by the example of

00:50:41.040 this abstraction so this is a application form class we do not inherit from anything that's a just a rubby

00:50:47.440 class but we include some active model uh functionality because active model is

00:50:52.799 actually what's used throughout all the rails parts to to bring you uh something

00:50:58.839 like attribute like behavior um and uh validations also part of

00:51:05.799 active model API and so on we add callbacks well yeah callbacks might be

00:51:11.559 evil but that's the Ra's way of doing things that's actually if you use

00:51:16.640 callbacks properly it is uh to separate

00:51:22.240 the main logic of your like function method class from the peripheral logic that the

00:51:29.440 best usage for callbacks and you should not avoid them for that so it's nice to have them actually we also have some

00:51:37.280 additional safety checks within this abstraction transaction awareness which

00:51:42.640 is important and uh yeah we have some interface just a single method we must Implement in your class to create any

00:51:49.000 kind of logic it's not necessary creating model it could just be send in email it's still a form submit finally

00:51:55.720 that's where we use model name it's actually used for active active now action sorry yeah still confusing uh

00:52:04.240 action view compatibility and that's uh one of the most important things here uh we're going to see it in action in a

00:52:10.559 second so let's see how using a rails W form object compares to using a regular

00:52:18.319 model so this is a kind of code using just regular rails model just a scaffold with some like alization added and this

00:52:25.400 is the code used in a for model object it stays almost the same anyone looking at this code understands what's going on

00:52:31.240 oh okay we create some form data we save it and we redirect that's how we do it

00:52:37.760 with regular rails nothing new what about the view so this is our form for

00:52:43.200 cable original one uh just to remind you it converts to this HTML like style

00:52:49.640 emitted and this is the same version for the form object we just changed the single word but everything stays the

00:52:56.920 same due to this model name thing we tell rails how actually to use this

00:53:04.200 object to generate URLs parameters and all the stuff that we use in various

00:53:10.160 action view helpers just by the matter of single method in the abstraction and

00:53:16.280 that makes our form object quack like a model so we don't need to change anything here so now it's time to actually

00:53:23.480 Implement our multistep form our wizard

00:53:36.799 so we introduced form object and that gave us actually enough flexibility to build

00:53:44.520 to and improve the functionality of this particular action operation to build on top of that because now we have a single

00:53:52.760 place in our code base responsible for this interaction and we can localize all

00:53:58.160 the changes if we want to change the user experience so we don't need to

00:54:03.640 touch multiple places out codebase to create a wizard we don't need to introduce any gems no create new

00:54:10.720 controllers nothing we can do everything but just relying on our form object

00:54:15.880 first of all what is wizard wizard is actually deterministic finite automaton yeah uh some nice pictures

00:54:23.839 anyway if you're curious you can ask me after afterwards uh it has some story behind

00:54:29.040 it um so what is a good way to implement uh automaton or state machines as many

00:54:37.480 people call it in uh software engineering of course we can use some State machine implementation and since

00:54:44.359 we're dealing with pure Ruby object we're not going to use any popular gems for State machines for active record

00:54:51.200 because they pollute your active record model and bring unnecessary complexity

00:54:56.359 we're going to use another gem uh I like it's called workflow it's it provides a

00:55:02.079 very minimal State machine functionality to any Ruby class you want and we're

00:55:07.720 going to use it like this so we create a separate class within our form we call

00:55:14.079 it Wizard and it's actually going to be a workflow and we Define State transitions yeah we use some DSL maybe

00:55:21.319 well at least it's it's readable and we can control the steps where to go from

00:55:28.200 where and which case because there's some conditional transitions right in this class and all

00:55:35.640 the logic of moving between pages is implemented here then we just use this

00:55:42.400 wizard to decide whether we need to really create the record or we just

00:55:48.640 need to step to the next iteration so we change our submit meod a bit

00:55:56.799 um again the code is kind of self-explaining let's see how it Chang

00:56:02.640 the way we write our controllers so for the controller we need to update just a single bit when we submit an

00:56:09.920 intermediate step we do not create the record so we do not redirect to the record page in this case so we must

00:56:16.640 proceed to the next form uh wizard stab and here we just respond with the same

00:56:23.039 new uh template but with a status created the reason why we use created

00:56:28.240 and not okay here is just to satisfy turbo because if you respond with okay

00:56:34.920 like 20 200 for post request it's going to tell you oh no you must redirect when

00:56:40.079 for submits so sorry I'm U I don't want to redirect but okay create it solve

00:56:45.480 this at least for now and if there is a validation error we just uh respond with a regular unprofitable entity and can

00:56:52.599 show errors right for this step the the most interesting part is here the only

00:56:59.039 other part of code that is changed is our form template that's what we do here

00:57:06.599 here we can ask our visard workflow what is the current state oh the current

00:57:12.359 state is name okay let's render the text fil name otherwise let's render a hidden

00:57:19.000 filed name that's how we persist the unsaved data okay let's go to the next

00:57:24.559 step same here we just add if and else and we render the current step or render

00:57:32.319 the state same goes again oh here we introduced something that wasn't in the

00:57:39.000 original version now when we have a m step logic we have some kind of a logic

00:57:44.440 tree we can actually add some intermediate steps and tell user okay you decided that you go with rails this

00:57:50.920 is some tips you need to know for using them any cable with rails go and run

00:57:56.240 these commands in your rails application so it's there is no input just just a um information step we go further we have

00:58:04.640 again this if else stuff finally one more step and at the final step we have

00:58:12.160 one input it's emitted from the template and we have some Logic for showing buttons here we again check for the

00:58:19.119 wizard State uh whether it can complete so what whether this step is potentially

00:58:24.839 final can lead to the final step State whether we can go back and we show the back button and we have some like HTML

00:58:31.960 magic using form action and value for the submit button to differentiate between back and forth operations and

00:58:39.039 that's pretty much it so that's all you need to introduce a mod step logic or whatever

00:58:45.160 step logic to your rail forms without changing your models changing your

00:58:51.480 controllers by describing your logic in a like human readable way as a state

00:58:57.039 machine and just a little bit updating the your template so that's all you need

00:59:03.200 if you create an obstruction that works the rails way yeah this one is

00:59:12.039 Happy well that's it for today actually I'm open to any questions because we

00:59:17.160 have a break right now I don't know if we're going to do it right now or we just stop the stream whatever so thank

00:59:33.839 hacking and one last thing I have some Fortune with me so if you want

00:59:41.240 to chase Your Luck come here and buite

00:59:51.079 one okay that's a 20 minute break 20 minute break I'm very loud um and bold

01:00:00.599 yeah um 20 minutes and then we uh meet for the open

01:00:09.520 mic thank you Rea can we all just give a big hand to Arena and evil martians for organizing this every

01:00:15.359 month before this year we haven't had a continuous Ruby meet up for many many

01:00:21.559 many years and I'm so glad to see so many old faces well sorry no offense but

01:00:28.039 like so many so many uh longtime friends let me put it that way and uh also New

01:00:33.760 Friends and it's just always re-energizing to come and see you all so I'm so happy and really grateful to

01:00:39.440 arann team for putting this together so big hand for them

01:00:45.599 please so my name is a pundit I'm up here every month I'm sorry thank you um

01:00:50.720 I'm the executive director of Ruby Central the us-based nonprofit that works to cultivate the Ruby Community

01:00:57.880 here in the United States and hopefully around the world if you're interested in following me on social media I don't recommend it it's mostly about bicycles

01:01:04.480 although sometimes so you can also follow Marco which is probably a lot more useful um and if you do follow me

01:01:12.559 you could see this kind of dumb stuff that I do while I'm listening to other people if you don't know Marco he's uh

01:01:19.359 at almost every single Ruby conference and he was saying that this year I think he's already done 12 Ruby conferences

01:01:26.079 and by the end of the year he's going to get to 14 maros a he is a true Gem and he's

01:01:33.079 working on all kinds of different projects including this Ruby video library he was just showing me uh which

01:01:38.280 is a mission to index all of the Ruby conferences this is pretty if you get a chance you should check this out um it's

01:01:45.440 done by Speaker uh conference um and even uh some of the sponsors so this is

01:01:51.440 an incredible project and I think we're all um this is the kind of stuff that we at Ruby Central are excited to support

01:01:57.240 and promote um because if you're here in this room there's only two reasons you

01:02:02.760 love programming in Ruby or and you enjoy being part of the Ruby Community

01:02:08.960 right either online or in person at the conferences and those are the two reasons why we're here and uh our

01:02:15.279 mission is to grow that conf grow that community and and help it be strong and

01:02:20.440 live on into the future as long as possible so check out Ruby video so again I work at Ruby Central um this is

01:02:27.400 our uh out ofate website um please help us update it if you have the opportunity

01:02:33.359 uh if you're interested in supporting us there's a lot of different ways to do that there's a link up here for support

01:02:38.400 us um if you want to become an individual member and support us financially uh you can go up in the

01:02:44.079 corner and click uh sign in and then there's there's options for um helping us financially which would be really

01:02:50.200 useful and helpful we do a lot of cool stuff and you can see all of our news there uh recently the big news that we

01:02:57.160 have is we welcome back Marty hot who ran Ruby Central for a very long time as our um interim lead for Ruby jems and

01:03:04.319 bundler in our open source program which is super cool so if you don't know about Ruby Central we admin we do two things

01:03:10.599 we do open source uh infrastructure and uh run events so you may have heard of

01:03:16.880 this program raise your again raise your hand if you've used this program great so the uh and then there's another uh

01:03:23.760 service and program that we run your hand if you use this one anybody heard of this website so we help keep this up

01:03:30.839 to date um we have recently completed a strict security audit so this service is

01:03:36.000 much more secure than it ever has been in the past and none of you uh probably know about that but we're going to be

01:03:42.279 working on putting a lot of that information out in the next couple weeks and helping you know like all the stuff

01:03:47.319 that's going on in the background to help your Ruby experience be um more secure first and foremost but also um a

01:03:55.079 better one we're developing organization level um administrative accounts for people who

01:04:00.839 are publishing many uh libraries that's coming soon and we'll launch that soon so that's exciting too um we also run

01:04:08.520 Ruby API uh or support and administer Ruby API which I recommend um and the

01:04:14.240 Ruby toolbox so if use any of these Services these are all supported by Ruby Central so we help keep them up to date

01:04:22.039 we make sure that they're secure um and we administer The Host and CDN bills and

01:04:27.359 things like that we have a lot of friends in the Ruby Community who we work with and support including wnb uh

01:04:35.440 the virtual community for women and non-binary rubis they have events at almost every Ruby Community event

01:04:42.760 conference um if this is a group that you identify with I strongly recommend

01:04:48.079 you join they're really awesome we have a light collaboration right now and we're working on making that collaboration a lot stronger so the

01:04:55.400 reason I'm up here is to invite you to Ruby Chicago um it's November 13th to

01:05:01.279 15th uh in beautiful downtown Chicago it is at the Hilton which is a gorgeous Art

01:05:08.240 Deco Hotel uh right downtown on the park it is the site of the 1968 um Democratic

01:05:15.200 National Convention and Associated riots um it's a really cool place it's Central

01:05:21.960 to the city and it's a great venue I strongly recommend getting your ticket soon and we have new

01:05:28.240 swag which is actually pretty fun um so

01:05:33.400 the shop is and this is Shopify a ruby and rails app uh store. rubby

01:05:39.240 central.org and I'm modeling for you right now the new uh swag for Ruby conf

01:05:44.279 so this is the zipup hoodie which just got released with the dragon on the back so even if you're not coming even though

01:05:50.400 you should come and you want to get a hoodie or some other rubycon swag uh you

01:05:55.640 you can pick it up there but again Ruby conf is uh November 13th to 15th in

01:06:01.240 Chicago uh and I would love to see everybody here attend uh there's some speakers here as well so if anybody has

01:06:09.039 any trouble getting to the conference please come and talk to me come message me we want as many people there as

01:06:14.480 possible um and one of the reasons that we want as many people there as possible is that conferences and Community only

01:06:20.760 exist if you participate so if you show up and the sponsors show up and then we

01:06:25.799 get to have our events and continue to be in person but you know as you see

01:06:31.599 from Ruby fuza which I've never attended but I know a lot of people have said really wonderful things about it's the

01:06:36.960 premier one of the Premier African Ruby conferences they weren able to get sponsorships and or tickets and so the

01:06:43.640 macroeconomic trends do affect our conferences and so if you want events like this to continue to exist encourage

01:06:50.160 your employers to sponsor and buy a ticket and invite a friend um and get a hoodie thank you

01:07:05.480 great thanks aders Serge your

01:07:13.920 next if anybody wants to buy it while I'm talking you're already selling a package

01:07:21.160 service come on you didn't even start to talk yet okay

01:07:26.720 um okay the hardest part is to like the microphone works now to figure out how

01:07:31.880 to share the whole screen and not to show something that is not supposed to be

01:07:38.760 there okay okay oh hi hi everyone so my name is Sergey Sergey I was the host of

01:07:46.599 U Yuko last year in vus uh I don't know who who have you

01:07:51.799 heard about Yuko uh like keep your hand up if you've

01:07:56.880 been to any okay cool awesome thank you guys so uh we also after the euruko

01:08:05.400 uh left to the other City we kick it off Baltic Ruby conference that's the conference that travel across the Baltic

01:08:12.599 seal uh this year we had it in uh Sweden in Malmo uh we had a couple of funny photos

01:08:19.319 of Ms but I don't have time to show it like if anybody's interested I will I

01:08:24.600 will show privately so this year Baltic Ruby comes to Ria lattia we're going to

01:08:32.120 have three tracks we have Tim Riley as a confirmed sponsor uh speaker as a

01:08:37.679 keynote and one of the track that we're going to have like actually we have one for Ruby one for open source and one for

01:08:45.239 everything else but not for rails like no offense but like I mean like

01:08:50.839 everybody's talking rails now and we we think like for this year maybe we need to give more like a spotlight for other

01:08:58.759 Frameworks like Hanami like dryer B and others so we can of encourage people who

01:09:04.199 have like who who interest in to come there but I wanted to show you one thing

01:09:09.359 that uh recently happened uh literally um we released the the gem

01:09:17.279 last week uh which is called rspec Lama so I'm interested who of you using

01:09:23.920 rpec okay who know who knows uh who created our

01:09:30.600 spec so yeah so the dude who created arpec called Steven Baker and he hates

01:09:36.759 arpec but we convince him to work together and make like a simple plugin

01:09:41.799 for arpec to kind of use arpec matchers and assertions to test uh local and not

01:09:49.319 only uh AI models so let me show you how it works so it's it's kind of pretty

01:09:58.600 um pretty easy so what do you need to do

01:10:03.679 um you need to add yeah it's going to be like I can handle okay so yeah you need

01:10:10.280 to add um now this one yay this one so you need

01:10:17.719 to add this Gem and in order if you want to use like any platforms some

01:10:23.199 credentials and and as soon as you have it uh and here is like a simple example

01:10:31.719 so let's have two classes so the one that's a pirate and uh it usually gives

01:10:38.000 you like a pirate speech so it says something like a pirate and another one

01:10:43.239 is like a robot so and like robots and Pirates they speak differently so if we

01:10:49.920 use rpec we kind of can validate that they both speak but our back cannot say

01:10:56.400 like does this guy speak as a pirate and this one speaks as a robot so using this

01:11:05.159 plugin for rpec you can easily do that so what do you do like you use the standard

01:11:12.120 uh uh standard syntax of rpec and you can configure which model you want to

01:11:17.800 run some parameters for the model and actually then check the uh the assertion

01:11:25.400 so does robot sounds like a robot and uh does the pirate sound like a pirate so

01:11:33.520 let's see uh if it's going to work or not because you never

01:11:47.320 open Ai and T what do we have here so it it

01:11:53.600 actually um shows you the output from the model so this one Yes sounds like a

01:12:00.080 robot but this one it says absolutely but we waiting for yes so it's a failure

01:12:06.199 which is let's run it again okay it's supposed to okay yes so

01:12:16.920 now yeah now like that's I think it's even better when you testing like using our spec and you use this

01:12:23.360 non-deterministic thing for uh for your unit test but uh it's not

01:12:29.880 only about using uh open Ai and the interfaces of open AI um arpec llama

01:12:37.199 also supports all the open not open like llama families like meta's llama

01:12:43.920 families and not only so we use local inference engine which is called llama CPP and bindings for for Ruby that exist

01:12:51.960 so for example if you don't want to use uh open AI uh you can download the

01:12:58.280 binary file of let's say metas Lama or whatever like MMO or mistal or other you

01:13:05.480 can specify in the in the line where is the model uh the actual binary file for the

01:13:11.560 model so like this one in dgf format so it's like literally something that lives

01:13:17.199 on your machine and um for this one let's ask more complex question so I

01:13:24.000 have this open Lama 7B uh download and uh let's ask who's

01:13:30.000 the creator of Ruby and as assertion we expect to be yukihiro Moto

01:13:35.920 son um so again let's see so it instantiates

01:13:42.880 the the local uh llama inference with llama CPP so there is no GPU but it uses

01:13:50.159 M2 CPU which is fine it's a little bit slower but so what do we have here who

01:13:56.480 is the creator of Ruby it gives a little bit more useless

01:14:01.639 information but finally as you can see it gives the right answer and the tests

01:14:07.120 pass so for example if you do you want to give it another try so for example we can ask another

01:14:15.000 question um you have any idea let's do rails right who's a

01:14:21.280 creator of rails B no no no it's like it's to too Fresh

01:14:29.199 So the creator of rails and uh let's say

01:14:35.360 dhh I think it should be fine let's

01:14:41.400 see so while it's it's doing this stuff so you can actually here's a creator of

01:14:46.600 rails yeah okay so you see it's it knows it

01:14:53.840 let's yeah yeah you can figure it out right so it's you don't need to read it all through we all doing this for M for

01:15:01.960 machines uh DAV

01:15:13.239 it you think okay now it's finally it sticks

01:15:20.520 with this with with its opinion uh which is David so um using this like simple

01:15:25.800 tool set you can build like multiple chains so if you thank you in if you use for example like for some purposes you

01:15:32.199 use open AI for other purposes you can use uh like your local uh models that

01:15:38.840 can natively using native extension work with Ruby so um yeah it's quite a fun

01:15:45.080 and I encourage everybody to try so the gem published last week it has the the very

01:15:53.080 first versions of give some support contribute and thank you very much for uh yeah for having me here thank

01:16:02.719 you perfect uh K kith uh come over yep your

01:16:13.840 turn Okay so mine is sort of a quick oh sorry I keep forgetting that uh

01:16:20.920 so I just have a real quick thing uh I'm speaking at Ruby

01:16:27.239 um in character um and if yall are coming stop

01:16:33.920 by I don't know where it's at it's in the program somewhere um it's uh lessons

01:16:39.600 that I learned uh running sidekick at scale at the large company that I work

01:16:45.400 for um and then secondly if yall are hiring find me at ruby.

01:16:53.560 Social um and uh or here whatever uh interested to

01:17:00.159 see what's available um yeah RTO and all that uh anyway but yeah that's it that's the

01:17:15.159 C thank you okay qu number four

01:17:22.679 um where is this yeah go yeah

01:17:29.040 Europe hello thanks for all the great talks um I'm Meredith I have been job

01:17:36.199 searching for far too long so um I came to the Ruby Meetup uh last month and

01:17:45.080 learned like 100% more about Ruby than I knew when I walked in um I was uh

01:17:52.560 formerly like devops SRE and then um yeah just looking for um any software

01:18:01.440 roles or Tech adjacent and um yeah I appreciate this community so nice

01:18:07.800 meeting you perfect uh so who has a different

01:18:16.679 Ruby roles please let merus know know about

01:18:22.080 them awesome all right um who's number five to be

01:18:27.520 honest maybe you ready yeah go

01:18:32.719 ahead hi uh kind of nervous to be here um actually I'm working on a product uh

01:18:40.280 a new test coverage service that combines both the CI simple Cup score

01:18:46.000 along with the production uh the production runtime heat map to tell you the actual test

01:18:53.360 coverage what I called usage weed test coverage so if you guys are interested in trying out it's currently very

01:18:59.840 unstable uh but I already published a gem called Oren cuff uh if you guys want

01:19:04.920 to try it out come talk to me I'll give you my name card um yeah that's

01:19:11.719 it wait wait say the name of the gem again uh oen o y n

01:19:19.760 coov yeah perfect thank you um we had jie it was you yeah right

01:19:31.199 right hello hello uh I'm Justin Bowen and I am

01:19:39.080 working on releasing an open-source gem so that we can build rails AI

01:19:46.880 apps and I'm also a Avid rubyist I've been using Ruby

01:19:53.760 on Rails for seven years but the last decade I have spent far too much time writing Python and the last year I've

01:20:01.320 actually had uh the opportunity to work with nextjs and part of the reason is because

01:20:08.440 we lack the tooling that those communities do I gave a talk at Ruby

01:20:13.600 comp 2022 on this topic which mostly mirrored what Adar just did with Marco

01:20:21.239 except with Andrew Kane if you're familiar very prolific rubyist who's made many gems including a lot of

01:20:29.239 machine learning and AI gems so I'm working on an initiative to

01:20:35.960 build a ruby AI working group and if you're interested in Ruby I hope you are

01:20:44.000 and Ai and you want to contribute to our open- source tooling come talk to

01:20:49.920 me thank you and if you're interested in the gem

01:20:55.480 I'm building active agents. forgot the plug and if you want

01:21:00.560 a sticker I have a couple yeah I I can recommend those stickers they're cool um all right um so

01:21:09.520 I think we're good with the open mic just to final second if

01:21:15.520 um somebody wants to go up but yeah I I think we're pretty good so

01:21:24.120 um with that Brandon go ahead start setting up um but yeah with that we are

01:21:32.000 um finally ready for the final presentation for

01:21:37.440 today um Brandon whoer is a I think principal software developer

01:21:44.320 at one medical and um we're going to be talking

01:21:53.080 about I hope going to be talking about

01:21:59.199 Lamor maybe some functional programming as well we'll see we'll see where it goes

01:22:05.520 but yep all right nice uh Brendon take it

01:22:16.199 away looks pretty good to me oh okay maybe too much Zoom yeah

01:22:24.880 uh what it is is I have to rehearse slideshow which will then put it there which does actually work no it's it's

01:22:31.880 good short version is zoom does not like showing my keynote notes and unfortunately I do not memorize things

01:22:38.320 very well so I need those notes so I had to make it work through this uh hacky if you want to see it's actually very

01:22:44.520 fascinating it's like literally please just zoom in on this narrow portion of the screen and I hope this works so here

01:22:50.760 we go so this talk is going to be the first

01:22:57.360 about 20 minutes of the Ruby keynote I'm going to be giving this year at Ruby comp I'm going to be on the third day

01:23:02.600 and the opening keynote for that day so I'd love it if you have any feedback any

01:23:08.840 type of ideas or otherwise but with that said we'll go ahead and get started with

01:23:14.920 red and fantasy land so first off who am I well simply

01:23:21.080 put my name is Brandon I love Ruby and I'm here to share that with you today and in fact I brought something special

01:23:26.600 a story I've wanted to tell for quite a while a story I deeply enjoyed writing and one I hope you enjoy as well with

01:23:33.120 that said should we go ahead and get started deep in

01:23:40.159 library we find ourselves in the Grand City libraries of the master Violet an

01:23:45.239 infinite well of information and wisdom in the world of the Lemurs why the library you might ask well you see

01:23:52.639 there's a particularly naughty lur named Indigo who's been repeatedly banned from the library and somehow keeps on finding

01:24:00.159 his way in along with his friend the esher cat and they're looking for something special in the deepest parts

01:24:06.960 of the library something so special in fact that we find ourselves journeying deeper and deeper into the depths of

01:24:13.280 Library where light begins to fade and a pressure can be felt Beyond even the depths of Lord Crimson is grimoires

01:24:18.960 there exists a book and to open that book is to incur a high cost indeed

01:24:26.040 so of course naturally we find Indigo and the eer cat retrieving such mysterious Blue Book one sealed with

01:24:33.000 ancient Magics and binding spells locked tightly in Chains and on its cover we can faintly make out the text Wonder

01:24:42.400 World upon opening the book of Blue Fog emerges the air in the room stirs and all the lanterns extinguish there deep

01:24:49.239 Magics here in Indigo and eer conspire amongst themselves to teach these Magics to a particularly unsuspecting young

01:24:55.320 lemur Indigo cackles to himself delighted discarding all warning notes and warning spells as he exclaims this

01:25:02.520 is perfect but they hear a noise a soft thump of slippers a light coming from

01:25:08.679 around the corner the master Library Violet was awake she enters the deeper Library

01:25:14.040 Lantern in hand trying to figure out what in the world is going on and why is so many alarms are ringing and found

01:25:20.000 Indigo in the eer cap with a particularly dangerous book what exactly are you Violet was cut off aser aser

01:25:27.400 time to make an exit whispered indigo violet and noid yells out after them Indigo aser don't you dare we cannot

01:25:33.600 have another eval incident especially not another mess with metaprogramming so mischievous smile came across the face

01:25:39.400 of indigo and the aser cat and with a snap of the aser cat's paw the two began to make their miraculous getaway fading

01:25:46.000 into dust Violet sigh and slowly approaches trying to find out which book had been

01:25:52.679 stolen and goes to look where the book once was seals undone multiple months

01:25:58.000 worth of binding spells broken the one book which must never be opened the one subject which must never be taught that

01:26:04.520 will render its reader incapable of ever teaching that subject again a subject no

01:26:09.679 wise lemur would ever dare touch so Violet takes a moment collecting herself

01:26:15.159 as the Shadows behind her stir and Coes and behind her several figures with

01:26:20.560 hoods drawn over their faces and Lead figures with a shining Stone Lambda pendant upon their chest the true owners

01:26:27.840 of the book they point to the missing book and the head liemer says simply it

01:26:33.199 must be found for there are some things the Lemur of this land are not ready for

01:26:38.600 something which must not be Unleashed within its Pages B agrees to this and says yes we can only hope that it can be

01:26:46.080 retrieved before that thing is let loose upon the world an ancient Creature From Time

01:26:52.719 immaterial granted his wisdom terrifying its potential who cannot be comprehended by the normal

01:26:58.560 lemur the only question left was where in the world Indigo might unleash such a force and what if anything could be done

01:27:06.239 to stop him in the eer cat the only hope was that no one would be caught up in

01:27:12.119 such a catastrophe and caught

01:27:17.800 unaware which brings us to part one a new adventure speaking of unawares we find

01:27:23.880 our friend read lemur studying in his tiny Forest Grotto he's just finishing up his book searching for his next

01:27:29.040 alignment but he found himself returning to the classics he sighed and looked up

01:27:34.320 to see saffron the great tailor sprinting on through with bolts of silk over his shoulders yelling early early

01:27:39.600 early I must be early as he dashed by he heard him repeating about tea parties and early and a piece of paper slipped

01:27:45.600 out of his bundle and landed in front of red early early early why was he so obsessed with being early and what was

01:27:51.920 this about a tea party red was confused set down his book and began to chase after sephron wait wait where are you

01:27:58.320 going in such a hurry asked Fred so many outfits to make I need to get there early before the party starts anything

01:28:04.480 through VV VIPs yelled saffron as he blew on by what tea party asked red

01:28:11.040 check your invitation they insisted you be invited as guest of honor said saffron so red stopped and took a look

01:28:17.400 at the invitation and saw what appeared to be a scraw of crayon with arrows pointing at what looked like a staircase

01:28:24.320 and uh what was this even when he looked up saffron was gone and in front of him

01:28:30.159 was a giant tree with a hole in the center a suspiciously well-designed sign that read fantasy land was it a trap

01:28:38.040 probably did Red particularly care probably not there was far more fun in

01:28:43.480 exploring and pushing the limits than sitting still and he wasn't going to miss out on whatever in the world this

01:28:48.600 adventure was so he found himself climbing through labyrinths of stairs right side left

01:28:54.679 side upside downside which way was even the correct direction which was was even up he found himself more and more lost

01:29:00.400 but at the same time approaching something until he found himself at a Gateway a gateway to a strange new

01:29:09.960 Land part two the forest of functions upon exiting the strange

01:29:15.520 Labyrinth he finds himself in a forest unlike anything he's seen before and in that Forest he can see a fire in the

01:29:21.600 distance and a few familiar voices as red enters the forest he comes across a fire and hears a group chatting was

01:29:28.440 told that Indigo was up to something again you think we should stop him why it seems interesting so far and who

01:29:34.800 knows what else him and that cat came up with said Lord Crimson it appears we have a guest shall

01:29:40.800 we serve him some tea just uh not a party we shouldn't interrupt those for later said

01:29:46.639 lavender red emerged from the forest confused and asked what in the world is going on in this strange place the two

01:29:52.880 looked at each other and began to laugh and then Crimson answered it appears our friends Indigo and the eaat are having

01:29:58.719 their fun and they've been so kind as to involve us in their plans to get ready and also to supply the tea for the

01:30:05.480 little party said lavender red looked confused at this it's an odd place yes but we'd best

01:30:13.000 get you ready for what lies ahead young one we're here to talk to you about functions today but just a few and a

01:30:19.320 quick review for those who need more of an introduction others might know some of this already said l

01:30:25.760 now in Ruby there are a few types of functions the first two that you might see are prox and lambdas but we'll be

01:30:31.520 focusing on lambdas for today and you can find QR codes in the top right which will link you to just I'll eventually

01:30:38.199 have a repo for that so in fact there's a type of programming called functional

01:30:44.000 programming which composes smaller functions together to do larger tasks in this case we have two functions one

01:30:49.960 which adds five and one which multiplies by three in Ruby we can compose those

01:30:55.080 using the double angle bracket when used in this direction multiply 3 is called

01:31:00.119 first but if we flip it around the opposite would be true and that you get both of those functions still

01:31:07.760 called now there's even a particular odd function which does nothing why would

01:31:13.400 you want to do nothing well if we use it we'll notice that it does indeed quite literally do nothing

01:31:20.480 it takes in parameter V and returns that parameter a function after my my own heart I must say but very useful in

01:31:26.000 certain contexts we can even use it to prove if

01:31:31.080 you'd like but it does indeed do nothing such an odd little function but what I have a feeling you'll be seeing quite a

01:31:37.080 bit of later functions are everywhere in functional programming they take all types of unexpected shapes did you know

01:31:43.239 you're already using them in Ruby today you might recognize this and say of course well it's a block well yes we

01:31:50.639 can call them blocks but they're also functions block block function function the same difference really the important

01:31:56.800 part is what they do and in this case they allow us to select only the even numbers out of a

01:32:03.320 list if we wanted to we could even make our own method which took a block function itself and an argument like so

01:32:10.000 and what that would be doing is we'd be starting with what we call a predicate function you'll notice I don't call these blocks because I prefer to reflect

01:32:17.360 on what it's actually doing in this case we're using it to decide whether or not an item should be included in a new list

01:32:25.480 and in this case what it's doing is if that is in fact true we're creating an item here and we're going to

01:32:32.480 give it to a new array and we add that item to a new array which we will return at the end of the method giving us only

01:32:38.840 the values for which the predicate function returned a truth result but there's something interesting

01:32:44.159 in Ruby and something to keep in mind which is there are many many different ways to

01:32:49.719 do something which is more correct well long hand shorthand select filter it it

01:32:55.199 really depends doesn't it which answer is correct depends on the code base you're working in and the people you're working with aim first and foremost to

01:33:02.000 be understood and not clever beyond that there are many lessons from other languages some with

01:33:08.840 different philosophies and what you might find with Ruby all are valuable certainly but be careful not to become

01:33:14.639 dogmatic said lavender with all that said there are others to meet and lessons to be learned

01:33:21.440 ahead from Friends old and new alike said Crimson so red wandered on to the

01:33:27.800 next area you you think we should warn him about the lavender was cut off nah it'll

01:33:34.520 be more fun if they learn it themselves answer Crimson part three The mapable

01:33:46.880 Meadows so we find red entering the mapable Meadows a place filled with flowers and plants of all kinds

01:33:54.080 in there ahead he finds chartreuse his teacher from many years past that taught him the ways of map but this is a

01:33:59.760 different place at a different time ah it's been what now six years my

01:34:05.719 how time flies and the young ones grow and so too does our understanding of what we think we might already

01:34:11.920 understand what do you mean Master last we spoke you taught me about map but wasn't implementing it with reduce all

01:34:18.280 there really was to learn what is map do you suppose

01:34:25.040 well of course it's using a block function to transform every element of a list and return a new list isn't

01:34:30.679 it yes that's true but there's far more to it let's take a look consider this example we know that

01:34:38.719 we can double every element of list by applying a function that takes a parameter V and returns back V * 2

01:34:45.480 but why do we know that how do we know that how do we even know what that block

01:34:50.520 function is going to do to that array how do we know that that trans Ates into a new array there are a lot of

01:34:56.639 assumptions we're making here but if we were to implement such a function for an array in that context certainly it does

01:35:03.520 do that and in doing so we notice that interestingly enough it returns a new array array in Array out if you will

01:35:10.920 that's because map in the context of Ruby's array means to return a new array with every value transformed by the

01:35:17.159 function we provided but what if this were a different type I'm not entirely sure

01:35:25.320 let's suppose that we had to Define map for a string what might that look like do you

01:35:31.320 suppose perhaps in the context of a string map would make a new string and

01:35:37.440 inside of that we' iterate through each one of the characters of that old string and then we would append the result of

01:35:44.000 calling a function on each one of those characters to append it to a new string which would be returned at the

01:35:50.159 end of the method and we would call it something like

01:35:56.320 this which gives us something very similar to what that map function was doing with

01:36:02.320 array but I think we can go further in our exploration What If instead we happen to

01:36:09.360 put map directly onto the string class much like what array has through

01:36:16.360 enumerable the noticeable differences here are that we're getting the characters directly from self and while

01:36:21.960 not required here it does make it clear where it comes from the other is that we're directly calling map on the string

01:36:28.320 itself there at the bottom which gives us something a little

01:36:33.480 bit more ergonomic but we could go further what if we were to implement map

01:36:40.320 using map there's certainly no rule against it and it would shorten our code a good deal too the two differences are

01:36:46.480 now that we're using map but we're also using join to join the characters back together at the end in fact we could

01:36:53.080 even take the one step further to condense that to a short hand to pass the function directly to the new map

01:36:58.920 we're calling on strings characters now that all said we should probably not monkey patch core Ruby classes as

01:37:04.960 generally frowned upon nowadays isn't it so what other options might we have here

01:37:10.520 well to start we could wrap our map function in a module like this and call it like so which would look similar to

01:37:15.840 our initial example the problem with this is let's say that you wanted to map more than

01:37:21.920 once It's Quickly going to become unwieldy so

01:37:27.040 we're going to need to find another option another option which allows us to express this idea in something a little

01:37:32.320 bit more containable What If instead we mapped our string in a class our own custom

01:37:39.040 context where we Define the rules and we get to decide what map means for that

01:37:44.119 context we can even Define map on it but this map has a problem do you see

01:37:50.040 it it returns back a string meaning that if you were to call map twice the code right here would crash

01:37:58.480 because string has no map method on it that's because we went outside of the context of our class and returned a

01:38:04.760 string instead so what's one to do well what we could do is take our code right here and we

01:38:12.400 can make a new method that we call WP which is how do we put it back inside of our container how do we put it back

01:38:18.400 inside of the context that we're dealing with and we can introduce the term R to do just that so every time we map we

01:38:25.400 would also make sure to use wrap right before we do so and it would work about as well as

01:38:34.360 you'd expect it to but here are the things to keep in mind the value is still in the context of our string characters class and we need to get it

01:38:40.840 back out having our own rules is great but our program probably wants a string it has no idea how to deal with string

01:38:45.960 characters so how do we get it back out well we do that by calling

01:38:51.599 value which gives us back that result on the other end so let's take a look back at our

01:38:57.599 implementation it looks like we have some things which could be used elsewhere no like the idea of rap and the idea of a value what if we happen to

01:39:04.880 pull that into say a module and we call that module something like mappable to

01:39:10.560 represent something which can be mapped and follow certain set of rules this basic example for making our own context

01:39:16.760 but it does serve a point by doing so our implementation of string characters is shorten to this but

01:39:23.440 but what's more is that you'll notice this pattern is very common like let's say I didn't want string characters I

01:39:29.159 decided that I wanted to instead iterate a string by its words albeit a naive implementation but still we can now

01:39:35.880 transform that or let's say that I happen to want to instead even do the lines of a

01:39:41.239 strain these are very general shapes and they can apply to a lot of different things and by establishing our own rules

01:39:47.840 for map we can do a lot of things which may not make sense for classes like string or array we can Define our own

01:39:53.440 rules of what it means to map in our own context in our own way and we can even use that idea if we were to use

01:39:59.119 something completely different like let's say take this single white flower here could that also be

01:40:06.960 mappable in the context of a single flower we have a color inside of it and mapping in this context means to

01:40:12.719 transform the color and return a new flower we could even add rules to it defining what a valid flower is by

01:40:18.639 checking if the value is indeed a color an easy mistake to make here is to think that map only works on collections not

01:40:24.960 so map just means to transform and to return the same type and a flower is

01:40:30.360 just as much a context as an array is or string is or hashes or any other Ruby type might

01:40:36.520 be so to test this we might take white and red here and create a new Flower by which we Mak 0% white with 100% red

01:40:44.239 effectively painting this flower red now there was a song Once I'd heard

01:40:51.840 about painting flowers red but that's neither here nor there but now that we have an idea of

01:40:58.679 what it means to be mappable there are also a few rules we must keep in mind ones you might already have noticed the

01:41:04.639 reason for earlier the first rule is that if we do nothing it should unsurprisingly do

01:41:11.480 nothing we start by defining equality for a mappable module for something as mapable to be equal it should be both

01:41:17.719 the same type as well as have the same value inside of it now let's take an example of a string

01:41:24.440 like hey world over here we make ourselves a do nothing function which takes no parameter V and returns V and

01:41:31.480 we give ourselves a new string characters class now afterwards we can check if it

01:41:37.639 does nothing to the rapper and in fact it does do nothing and we get back true response that means that doing nothing

01:41:43.560 in the context of something which is mappable does well nothing you might wonder why in the

01:41:49.679 world do I care about doing nothing Well back with our module context with our string characters class before map these

01:41:56.520 would have failed if it fails that means we're not returning the same type and we can think this as a baseline to ensure

01:42:02.560 our mappable class works as expected rule number two is if we call

01:42:08.199 map twice using two different functions we should get back the same result as if we compose those two functions

01:42:13.719 independently so let's say that we had two functions one which wraps a character in Brackets and the next one

01:42:19.159 which gets the success of next character such that a do success is going to give us back

01:42:25.239 B we could again create our context as an experiment here with the string hey world and then map it twice using the

01:42:32.560 function separately and use the same thing by using functional composition

01:42:38.080 here why do we care about this because after all these are just functions we shouldn't be changing how functions work

01:42:43.800 they're just functions they should always behave as expected our string characters class should not be doing anything unexpected mutation extra steps

01:42:51.000 and this is a way to prove that we're doing exactly what one might expect what you see is what you get no

01:42:56.679 surprises we may be able to Define our own rules but for functions they have their own rules we should respect those

01:43:02.679 and we should be very careful not to break them and do unexpected things though in some cases in Ruby this

01:43:09.520 does not hold true such as Ruby's hash instance doesn't behave the way you'd

01:43:15.119 expect it to because inherits map from inumerable and returns an array instead

01:43:20.320 now we must be very careful of this but Ruby's hat also has another method which works far more like map which is a 2

01:43:27.400 method which actually takes in a block function as well and it behaves a lot more like what you expect now here's a

01:43:34.280 question does that mean that we should change map on hash probably not because

01:43:39.639 do you realize how much Cod relies on that and how much Behavior probably is built upon that presumption that would

01:43:44.920 be very dangerous so there's a reason to get this right and to follow rules in the initial set which makes it a lot

01:43:49.960 easier to build upon things but map map can do so much more than

01:43:56.119 what we've seen and in fact it can even reach into the future if given a

01:44:01.239 chance our context might even be the future itself it could contain a thread

01:44:06.520 which will eventually finish and what we can do inside of this is that we can take map and add the

01:44:14.000 context of making HTTP requests which will eventually return and we can eventually parse adjacent it and we can

01:44:19.639 eventually get the value this is exactly what you might see in it's a trick we learned from lemur of JavaScript which

01:44:25.199 is a very fascinating concept so you see red there's far more

01:44:31.599 to map than it appears an entire world of mappable things to discover out there in the wide world Ruby we've only

01:44:37.560 covered a small portion of its potential consider it a pattern to be used when you find things in your programs with

01:44:43.159 similar shapes and that follow those rules so with that red had learned more about map and began to walk to his next

01:44:51.040 destination which brings us to the the end of what I actually have prepared because unfortunately I have to draw all

01:45:07.239 time yep I think I'm about 90 hours worth of artwork in so far and I have about 24 left so I'll make it by the

01:45:13.320 conference but uh that's going to be chancy oh yes trust me I'm wor I have

01:45:19.400 myself a deadline of the 24th actually do I have that Chrome window open I might just be willing to show you

01:45:27.199 the magical sheet that I'm going to make sure to censor the names of the slide so it's to

01:45:33.960 not give you spoilers of things zoomy zoomi zoom zoom zoom where is the zoomy share there you are

01:45:42.840 screens share yep about 116 hours total with the commitment of going through

01:45:48.679 draft Blue Line ink color slides notes and various other things and uh uh

01:45:54.199 there's a lot of red that I'm not looking forward to this weekend but we will endeavor and we will get it done but those are all custom illustrations

01:46:00.520 drawn by hand and for me I like giving Illustrated stories to give a bit of a different experience from what people

01:46:06.000 are usually used to and I like to tell stories around functional programming or other novel Concepts in Ruby now granted

01:46:12.520 some of those are much harder than others and some of you might be guessing where I'm going with this and know exactly what type of danger I managed to

01:46:18.199 set my foot in so there are be traps there I'm GNA try my best to see if we can pull this talk off

01:46:23.480 but they said we want the most technical talk you can come up with you have an hour figure out what you want to do and that's the only chance I'm going to get

01:46:29.400 to give this type of talk so here we

01:46:43.679 questions we can we can take some um

01:47:13.080 uh if you paid attention as well whenever red was reading the book there is a why Fox on the front of that book

01:47:19.159 as well and I'll put a whole bunch more references in there this is the initial draft version which means I haven't

01:47:24.840 actually done all the finishing touches on this so if you pay attention during the live talk you're going to see in

01:47:30.199 library every single one of those books is going to have some sort of reference to some sort of Ruby book including ones by say uh Vlad and noell and I should

01:47:37.159 really write a book I feel like I'm the only person up here doesn't have a book I mean to be fair I did have

01:47:42.360 contract for I should probably finish it they're probably mad at me by now anyway uh moving on from that subject no I try

01:47:48.320 and sneak as many Easter eggs into there as possible like you might have noticed during the staircase there's a special

01:47:53.880 little uh time and space vehicle which happens to be flying around in the background

01:47:59.800 there technically it is the dartis which is day and relative time in Ruby or

01:48:05.800 something like that I don't forget what exactly the acronym was but I had a doctor who variant talk which was going to be at rails 2021 but uh certain Coy

01:48:13.880 events happened and I didn't finish at once so I try and make references to it so I can eventually give that talk but

01:48:19.080 it's like half done I should probably finish it

01:48:25.800 uh do we have any other questions or other things I appreciate that's a very very heavy subject and can lose the heck

01:48:31.480 out of everyone so I try to make it as approachable as I can but it's rough to figure that one out so a lot more build

01:48:37.760 steps just takes a bit go for it

01:49:19.520 mhm yep yep so there's a fun little secret in the end of the talk to where I almost can't do that so the way it works

01:49:27.719 is that I'm hiding something in the middle of there and I'm trying to introduce Concepts in a way which is approachable at least decently to Ruby

01:49:34.440 people but also not telling people what I'm actually doing because the second I say those words out loud everyone's going to lock up and say oh my gosh I'm

01:49:40.960 learning about X I cannot learn about X I've heard that's hard and that's impossible so there are certain words I

01:49:46.440 can't say out loud quite yet which happens to relate to the Beast of the book and the symbols on the back

01:49:53.840 of the hoods of all those L that showed up in the library but no it's a tough line to run

01:50:01.360 between I want to tell you but I shouldn't tell you but I want to but I shouldn't so I'll try and find a better way to approach

01:50:11.440 that go for

01:50:19.440 it oh it's going exactly where you think it's going and it's exactly I'll put it this way whenever I said it's a subject

01:50:25.800 which must not be taught because if you ever read the book and understand the book you're not allowed to teach it

01:50:30.920 anymore because you'll never know what in the world to explain it

01:50:40.400 with now the trick is and I'm not sure I know how to explain it either so let's hope I figure it out between now and

01:50:47.360 then I keep on doing this to myself I give myself an impossible challenge and say what happens if I try and teach this

01:50:52.440 thing I have no business teaching I guess I learn a lot one time I even showed up with a BTO

01:50:58.199 looking wig and tried to learn how in the world to do it Symphony Orchestra using midi with the audience's cell phones that blew up on stage but hey it

01:51:04.320 was

01:51:16.679 right all right this um this concludes our event today but we are planning

01:51:23.599 to go check out this cafet turus like a bar super close by um so

01:51:33.199 you're all welcome to ter ter ter tuse right something like that um I put

01:51:40.920 it on Luma by the way uh the name of the the place it's straight here like around

01:51:46.599 the corner and yeah just want to say thank you to

01:51:53.000 uh to our hosts at Chim uh again for for hosting us today so thank you guys and

01:51:59.119 let's let's just give them a shout out and I'll see you next time uh yep

Sections in this recording