text
stringlengths
0
1.82k
• Using query string parameters to determine the view to use
• Translating the chosen view into Swagger
• Different responses for an action
• Documentation of multiple media types is simplified and beneficial
• Concept of views having different representations for various use cases
• Simplification of abstraction in the DSL (Domain Specific Language)
• Discussion of upcoming projects, news, and future functionality for Goa
• Plans to finish up security examples
• Releasing a stable version of Goa (1.0)
• Moving on to VNEX 2.0 development
• Exploring extensions beyond HTTP, specifically GRPC
• Addressing abstractions that don't match HTTP abstractions
• Writing plugins and DSLs for a programming language
• Defining own output for plugins and built-in generators
• Difficulty in modifying built-in generator output for low-level HTTP server glue
• Making the language more open and allowing contributions from others through plugins
• Discussion of Goa design and Slack channel for collaboration
• Upcoming conferences: abstractions and GopherCon
• Discount code "GOTIME" for $50 off both conferences
• Raphael will speak at GopherCon, speaker on Goa topic
• Organizer of abstractions conference mentioned
• CLI tool (MK) discussed as ideal for its clarity and examples
• Comparison of Cobra and Viper
• Ease of use and understanding of Viper
• Documentation and integrations of Viper
• Defining flags as slices or maps in Viper
• Discussion of a blog post on application data caching
• Data storage and structure
• REND project: open-source tool for data storage and compatibility with Memcache D and RocksDB
• Use of RocksDB as an L2 cache to reduce memory costs and financial expenses on Amazon instances
• Examples of companies using RocksDB, including the REND project, CockroachDB, and others
• Discussion of RocksDB and its origin from Facebook
• Shout-out to Scott and the Netflix team for a thorough write-up on their Go proxy using RocksDB
• Performance metrics of the Go proxy, including handling of 2 million requests per second
• Overview of Shield, a tool from Stark and Wayne that can be used as a universal utility knife for backing up systems, with Dr. Nick from the Ruby world endorsing it
• Backing up databases and disks
• Shield tool for backups
• Hekka backup system comparison to Shield
• Zap structured logging framework from Uber
• Leveled loggers and structured logging systems
• Discussion of an efficient and feature-rich tool for distributed queues
• Recap of the show's closing tradition to thank open-source project contributors
• Brian's recommendation of NSQ from Bitly as a favorite open-source tool
• Description of NSQ's benefits, including its speed, predictability, and reliability
• Mention of Matt Richardson's talk on NSQ at Go4Con 2014
• iTerm2 is recommended for its new features and non-intrusive tips
• The latest beta versions of iTerm2 have added radical features and toys
• The speaker uses a Linux workstation and compares it to iTerm2
• Rethink DB has been used by the team and found interesting
• The speaker discusses the feature set of RethinkDB and how it fits with their use case for generating events when data is updated.
• The built-in subscription feature in RethinkDB has changed the way they think about designing systems for new services.
• The speaker recommends taking a look at RethinkDB for its capabilities and another dimension it adds to system design.
• Open-sourcing projects like RethinkDB is encouraged, as it benefits both employees and companies.
• The speaker praises companies that allow their employees to develop open-source projects.
• Discussion about cheating in a game
• Removing a score from Eric's scoreboard due to cheating
• Topic of log structured merge trees and their functionality
• Appreciation for Rafael and his expertise on code generation
• Wrap-up and thanks to the audience and special guests
• Goodbyes repeated multiple times
[0.00 --> 2.46] I'm Rafael Simon, and this is Go Time.
[17.46 --> 22.18] It's Go Time, a weekly podcast where we discuss interesting topics around the Go programming
[22.18 --> 27.06] language, the community, and everything in between. If you currently write Go or aspire to,
[27.06 --> 28.62] this is the show for you.
[30.00 --> 35.94] All right, everybody. Welcome back for another episode of Go Time. It is episode number seven.
[36.68 --> 40.52] Today, we have Brian Kettleson here with us. Say hello, Brian.
[41.06 --> 41.44] Hello.
[42.18 --> 44.26] And Carlisa Campos is also here.
[45.04 --> 46.38] Glad to be here. Hi, everybody.
[46.88 --> 54.26] And we also have a special guest here with us, Rafael Simon, who is the creator of a framework
[54.26 --> 60.38] called Goa for generating APIs, which Brian is particularly excited about.
[60.94 --> 61.20] Hello.
[61.20 --> 65.36] You want to give us a little bit of background, Rafael?
[66.28 --> 67.14] Sure. Yeah.
[67.34 --> 70.66] So let's start with who am I.
[71.24 --> 74.12] So I'm a platform architect at RedScale.
[74.64 --> 76.90] RedScale is a cloud management platform.
[76.90 --> 80.88] I've been working there for almost eight years.
[81.22 --> 84.84] And when I started, the whole product was basically a single raise app.
[85.48 --> 87.66] And the platform has grown a lot since then.
[88.00 --> 92.12] And last time I counted, there were about 52 different services running in production,
[92.80 --> 94.42] you know, running on about a thousand VM.
[94.42 --> 99.56] So I've helped design, develop, and debug a lot of them.
[100.96 --> 106.96] And part of going from this single raise app to all those distributed services,
[107.54 --> 113.58] we felt a lot of pain in having to design APIs the right way.
[114.14 --> 120.26] What I mean by that is being able to come up with APIs that are consistent
[120.26 --> 124.56] and have standards that are enforceable.
[124.86 --> 127.24] So, you know, so that we can come up and say,
[127.38 --> 128.90] yep, that API looks good.
[129.24 --> 130.42] It follows our standards.
[131.10 --> 135.06] And we'd be able to integrate that service with the rest of the fleet.
[136.06 --> 139.70] And as you probably know, once an API is alive,
[139.70 --> 141.96] it's almost impossible to change it.
[142.28 --> 144.32] Once you have customers that start using it,
[144.42 --> 147.54] or once your internal services rely on it,
[147.54 --> 151.52] then that API is going to be alive forever.
[151.78 --> 155.46] So it is very important that you spend the time designing it properly.
[157.42 --> 160.84] And when we looked at what was available to do that,
[161.28 --> 162.34] there just wasn't much.
[163.24 --> 166.08] There were a few tools here and there,