text stringlengths 0 1.82k |
|---|
• Discussion about the ease of building distributed systems using technologies like etcd and RocksDB |
• Discussion of GPUdb and its status as possibly "vaporware" |
• Introduction to Syncthing, a peer-to-peer syncing tool written in Go |
• Features and benefits of using Syncthing for syncing files across machines |
• Mention of alternative methods for file syncing compared to Dropbox |
• Discussion of Caddy, a web server/reverse proxy with built-in Let's Encrypt support |
• Caddy's embeddability and potential for programmatic configuration |
• Introduction to Viper, a command-line tool for pulling in environment variables and config files |
• Use cases and features of Viper |
• Discussion of the Cobra library and its benefits for writing command line tools in Go |
• Mention of Steve's Hugo project as a precursor to Cobra and other libraries such as Viper and Pflag |
• Introduction of Gin Web Framework/ API framework, which simplifies web development in Go |
• Plug for goa, a code generation tool that creates API implementations from Go DSLs |
• Discussion of the benefits of using goa, including readability and the generation of Swagger JSON files for documentation |
• Clarification on how goa works, including defining APIs in Go code with anonymous functions |
• Discussion about Go database access layers, specifically GORM |
• Introduction of GORMA plugin for Goa framework |
• History of using Go at Iron.io, including adoption before 1.0 release |
• Comparison with other languages like Java and Ruby |
• Role of early adopters in promoting the language's growth |
• Importance of community passion and interaction in driving interest in the language |
• Productivity with C++ vs Go |
• Go's concise syntax and small number of keywords |
• Lack of code generation needs in Go development |
• Early pain points with Go (library availability) |
• Comparison to other languages like Java and Ruby |
• Performance optimization in the Go community |
• Allocations and memory management in Go programming |
• Discussion on static analysis tools for Go programming language |
• Importance of memory allocation management at high scale (1M requests/sec) |
• Challenges of tracing down allocations in large codebases |
• Hiring challenges for Go developers and experience with training new hires |
• Adoption of containers, specifically Docker, by the company |
• Use of Docker to enable easy testing and deployment of customer code |
• Discussion of Docker and containerization |
• Experience with early bugs in Docker and how they were handled |
• Advantages of using containers in development and production environments |
• Future prediction for Docker and containerization taking over the computing world |
• Shoutouts to open-source projects, including GopherJS, polymer bindings, and gocyclo |
• Docker and containerization |
• fsnotify library for monitoring file system changes |
• Upcoming major announcement by Travis' company |
• GoTime.fm podcast and newsletter subscription information |
• GitHub suggestions for show topics or guests |
• Appreciation for Travis Reeder's contributions to the Go community |
**Erik St. Martin:** It's Go Time! This is a weekly podcast featuring special guests where we will discuss interesting topics around the Go programming language, to community and everything in between. If you currently write Go or aspire to, this is the show for you. |
Okay, so episode number three. Today we have Brian on the call, why don't you just tell everybody hello, Brian? |
**Brian Ketelsen:** Hello. |
**Erik St. Martin:** And we also have Carlisia here as usual. |
**Carlisia Thompson:** Hello everybody. |
**Erik St. Martin:** And we have a special guest today, long time Go community member - and I mean _long_ time Go community member - he's got a great beard, and he's also the CTO and co-founder of Iron.io. We have Travis Reeder here, tell everybody hello, Travis. |
**Travis Reeder:** Hello. |
**Erik St. Martin:** Typically we start the show off by just talking about any news and articles that we've come across. Brian, Carlisia, do you have anything? |
**Brian Ketelsen:** Oh, I've got something big. We all follow compiler times on TIP these days, because they can either... |
**Erik St. Martin:** No way, nobody does that. |
**Brian Ketelsen:** Yeah... The compiler slowed down a little bit in Go 1.6, and I saw a tweet from Dave Cheney recently that showed that one of the most recent commits cut compile times - I'm looking at his [juju](https://github.com/juju/juju) graph, it looks to me like it cut about 40%, so we're getting much closer b... |
**Erik St. Martin:** Yeah, I think that that was kind of a public thing, when they converted the compiler to Go, because a lot of it was kind of done through code generation that we all knew that that would happen. But it's great to see the performance come back. Is that part of 1.7? Is that locked into that release, o... |
**Brian Ketelsen:** There's not a feature freeze for 1.7 yet, so I'm assuming that anything that's in TIP now is a candidate for 1.7 and there's no reason it wouldn't be included, unless it breaks everything. |
**Erik St. Martin:** I'm super excited about that. Compile times what - doubled? It was something along those lines. |
**Brian Ketelsen:** At least doubled, yeah. They were harsh. |
**Erik St. Martin:** It's still nothing compared to C or C++, but we have to have a reason to hate, right? |
**Brian Ketelsen:** When you have lightning fast compile times and they go down to just fast compile times, everybody whines. |
**Erik St. Martin:** \[laughs\] The fast isn't fast enough. |
**Brian Ketelsen:** The nice news about these last couple releases is that although compile times went up a bit, performance has improved quite a bit, so with the ssa changes, I think Go is speeding up in general, and if we can get those compile times back down to where they were, we win on both sides - both sides of t... |
**Erik St. Martin:** Yeah, great. |
**Brian Ketelsen:** So I have another news item that I thought was interesting, but I am completely unqualified to talk about. There was a CloudFlare blog post about building the simplest Go static analysis tool possible, and it made me think that maybe ssa is something within my reach, and it was a very interesting re... |
**Erik St. Martin:** I've played with some of the static analysis tools that have already been created, but I haven't created any of my own. I did see that CloudFlare post though, and yes, that does make things look approachable. I'm not sure that I have anything in particular I want to write yet, but it does sound fun... |
**Carlisia Thompson:** New Relic has a static analysis tool, correct? But I don't think it has support for Go yet. |
**Erik St. Martin:** To be honest I haven't look at New Relic in a while. In my Ruby days there was a lot of New Relic, but I don't think I've used it with Go, so I'm not sure what support they have. |
**Carlisia Thompson:** Is it a general thing that people who are using Go are not using New Relic, I wonder? |
**Brian Ketelsen:** I think they just recently added some Go support, I don't know how much though. |
**Travis Reeder:** Well, it was certainly popular in the Ruby days, that's for sure. |
**Brian Ketelsen:** It was, everybody had New Relic. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.