text stringlengths 0 1.17k |
|---|
[1633.32 --> 1643.32] and um stop asking yeah i know it's like rtfm dude but um yeah so that that kind of did take up |
[1643.32 --> 1648.64] some time but since it was so small a lot of the changes just required thinking really carefully |
[1648.64 --> 1655.10] about kind of the trade-offs that's another thing about shell scripting is it seems simple on on its |
[1655.10 --> 1661.10] face but every single command every serve behavior has like a dozen or so edge cases you have to be |
[1661.10 --> 1667.58] cognizant of and then there's also implementation differences between the shells uh based on like |
[1667.58 --> 1674.72] what's how they evolved because like bash came from ash um and there's also dash which is the |
[1674.72 --> 1681.12] bin sh on dbian systems which is like super minimal it has barely anything like most bash code will not run |
[1681.12 --> 1689.56] on it um and then zh zsh actually came from ksh and they have their own weird bizarre features um that many |
[1689.56 --> 1694.46] people probably don't know about for instance arrays and zsh are index starting at one not zero |
[1694.46 --> 1700.00] so i can't really rely on that yeah i know they they thought it was a really cool feature that's |
[1700.00 --> 1705.80] stupid sorry yeah i know yeah but uh yeah the kind of way to get into that the more you have to like |
[1705.80 --> 1710.72] sit there and actually test and ponder and that's actually the other thing is i was really super |
[1710.72 --> 1717.04] aggressive about unit testing shell scripts and it always seems like when you when new languages |
[1717.04 --> 1724.40] come about that don't have like test suites or people kind of like misinterpret the language or |
[1724.40 --> 1729.32] like i don't know they think it's really simple they always like say oh we don't need a test why |
[1729.32 --> 1734.72] would you test that just just run it from the command line yeah you don't need unit tests here |
[1734.72 --> 1739.36] and people said this for like javascript when it was first starting out when like you know most code was |
[1739.36 --> 1745.22] just like three functions and then they also said about bash script but then once the testing tools get |
[1745.22 --> 1751.98] developed and people get you know used to them it really does help and that helped uh kill a lot |
[1751.98 --> 1758.50] of bugs and actually it was really cool because a lot of the other people who developed their own |
[1758.50 --> 1764.52] ruby switchers uh came in and started like uh suggesting features and you know how to solve |
[1764.52 --> 1770.74] various implement the auto switching so we actually were discussing this in issues by sending submitting |
[1770.74 --> 1777.02] pull requests for example unit tests so we really were just like speaking using the test as the |
[1777.02 --> 1781.94] implementation and so that's really cool to see uh instead of like you know getting these long-winded |
[1781.94 --> 1788.04] discussions and issues and we could just be like here's the code yeah and i can imagine that like |
[1788.04 --> 1792.30] looking through pull requests you know kind of as an aside here looking through pull requests and stuff |
[1792.30 --> 1799.64] for shell can kind of be kind of a disaster right somebody like submits all this code like yeah you |
[1799.64 --> 1805.58] have to understand it and like yeah everyone has their own weird styles yeah there is although um shell |
[1805.58 --> 1813.52] scripting does it's probably even more style obsessed than ruby is uh there's this one user who uh |
[1813.52 --> 1819.10] contributed came in and basically just like submit submitted this giant pull request that broke all the |
[1819.10 --> 1823.54] test and he's basically trying to rewrite it from scratch and like it's like whoa slow down let's |
[1823.54 --> 1830.62] take this one step at a time and basically kind of like he got me to um fix all the style issues |
[1830.62 --> 1835.48] and wow like for instance there's like very hardcore these style issues are not just like |
[1835.48 --> 1840.14] you're supposed to do it because it looks nice but there's actually safety reasons behind it |
[1840.14 --> 1845.14] for instance most people uh when they start doing shell script they always see uppercase uh variables |
[1845.14 --> 1852.08] so they just assume all variables are uppercase but all script local variables all function local |
[1852.08 --> 1857.72] variables are supposed to be lowercase so they don't overshadow the other global variables that |
[1857.72 --> 1863.50] are passed in as an environment variables quote um yeah so that's kind of like one of the major |
[1863.50 --> 1869.36] things we had to change and yeah there's other stuff that's crazy so awesome have you had any other |
[1869.36 --> 1874.90] projects that have gotten as popular in the open source community as these geez um |
[1874.90 --> 1879.34] trying to think here |
[1879.34 --> 1889.90] maybe maybe bundler audits that's that's probably it um or actually there's another um library i did where |
[1889.90 --> 1896.28] after ruby 1.9 came out ruby zip was kind of unmaintained for a long time it didn't work |
[1896.28 --> 1901.18] and so i just got frustrated and forked it and you know tried to get it working in 1.9 |
[1901.18 --> 1905.68] and released kind of another gem and apparently that got a lot of downloads because a lot of |
[1905.68 --> 1910.74] people are just like i just need this library to freaking work yeah well the reason i ask is because |
[1910.74 --> 1915.46] you know you're we were talking about like ctfs before the show and just like your attitude um |
[1915.46 --> 1923.92] seems you know you go by an alias and so it seems like privacy is a important thing to you i mean you're |
[1923.92 --> 1928.78] into security and all that so i wonder like was there any implications for you of like just getting |
[1928.78 --> 1935.34] popular like and getting like notoriety and getting pointed at for a lot of stuff well um i'm very |
[1935.34 --> 1942.08] like i've always been very anti-celebrity anti kind of cult of personality um coming from like the |
[1942.08 --> 1948.96] security hacking world uh it's it's a huge problem um like there's like you there's like you go through |
[1948.96 --> 1954.52] cycles where like someone will rise to fame and they'll be speaking at every conference and usually |
[1954.52 --> 1962.16] uh often recycling their you know previous talk and um and kind of that that community has developed |
[1962.16 --> 1968.20] this huge kind of like immune response to that where if you're not putting out um actual useful |
[1968.20 --> 1974.76] information new information new research um they quickly kind of like forget about you and sort of |
[1974.76 --> 1979.62] just like whatever like if you're just sort of being a thought leader uh you're not really as valued as |
[1979.62 --> 1985.58] much as someone who's actually doing like really useful work so yeah i guess um i'm very cognizant |
[1985.58 --> 1991.50] about not doing not doing that and one of the things i make sure to do is always give credits to |
[1991.50 --> 1999.84] the contributors and i also got a shout out for um uh the user havenwood on github has been a huge help |
[1999.84 --> 2006.92] in maintaining uh ch ruby has helped with uh always submitting um the homebrew uh rest updated homebrew |
[2006.92 --> 2012.14] recipes to homebrew and testing things on osx because i don't have an osx system currently so |
[2012.14 --> 2019.20] and it's like huge help so it's just not me being all like i'm not super genius here developing it like |
[2019.20 --> 2025.28] in a vacuum uh no yeah actually that's one thing that that kind of struck me as surprising is that |
[2025.28 --> 2031.14] ch ruby has 30 contributors on github um so obviously there's a number of contributors that actually |
[2031.14 --> 2035.92] probably not obviously but i would assume there's a number of them that have not actually uh you know |
[2035.92 --> 2040.08] contributed to the shell scripts themselves but to other things around them but but it's still i |
[2040.08 --> 2044.26] mean like it's 90 lines of code in the shell script and you have 30 contributors it's like that's a |
[2044.26 --> 2048.66] pretty big ratio and i think that's neat i think that's cool to see so many people like pitch in and |
[2048.66 --> 2052.62] help on different angles and you said like with the homebrew and you probably have people that have |
[2052.62 --> 2056.38] just updated the readme but people care about this and what you're doing and that seems pretty cool |
[2056.38 --> 2062.00] and also i think what came out of it was i wrote a generic make file because i was kind of annoyed |
[2062.00 --> 2068.14] of how all these shell script projects either they had this really simple install.sh file or they like |
[2068.14 --> 2072.92] you know you curl down and install script i just really wanted a simple make file they just installed |
[2072.92 --> 2078.72] worked on bsd and linux and like because there's issues with the version of make on bsd and linux uh |
[2078.72 --> 2084.54] the gene you make versus the bsd make and so a lot of those like there's like i guess you have a lot of |
[2084.54 --> 2088.90] casual contributors and they fix kind of the minor things i think that's really also important open |
[2088.90 --> 2094.92] source too um if you see a minor bug you should probably at least report it or fix it um because |
[2094.92 --> 2101.60] every little fix counts it does add up and even if it's like typos in the readme or you know this is |
[2101.60 --> 2106.42] a jumping forward but just on the i hate to put a lot on me because i don't like to do that whatsoever |
[2106.42 --> 2114.30] but you know you talk about little fixes that had a little fix in your public uh ronin um which was |
[2114.30 --> 2118.86] like because i was doing some research for the call and i was like oh this link is off but the |
[2118.86 --> 2124.10] coolest thing about github is that let me easily click edit fix the link for you and submit a pull |
[2124.10 --> 2130.10] request you know in a couple clicks and you know steve kladnik had an awesome post on the changelog |
[2130.10 --> 2134.08] i think about a year and a half back that still gets tons and tons of reads we'll link out to in the |
[2134.08 --> 2139.44] show notes if you're listening to this but you know like he had said all those little fixes 30 contributors |
[2139.44 --> 2144.78] they may not have done the heavy lifting but they're they're keeping a line on the edges you |
[2144.78 --> 2150.92] know that that you forget or you miss it's sort of the uh long tail yeah long tail principle where |
[2150.92 --> 2156.06] the casual contributors do actually add up over time and it's also kind of minor things do actually turn |
[2156.06 --> 2161.44] away a lot of potential users and so that's kind of when you do release code you have to be kind of |
[2161.44 --> 2167.30] ocd about that um if you see a typo in the readme they're probably gonna like they're gonna be biased |
[2167.30 --> 2172.00] automatically against the against the project um but yeah the funny thing about that typo was |
[2172.00 --> 2181.18] uh when github changed to all the user pages to like github io um i have this git alias get said |
[2181.18 --> 2187.44] so i can actually do mass find replaces and i end up breaking the url on every single page and so |
[2187.44 --> 2194.02] i had to undo that yeah i'm sure that was fun to fix so good really easy using get said but well |
[2194.02 --> 2198.18] it was a noise yeah i literally changed two characters which was awesome |
[2198.18 --> 2206.02] so going then to ronin i think ronin uh look at through all the projects that were listed i think |
[2206.02 --> 2211.86] ronin was the probably the first one uh well not the first one but one of the bigger ones that i've |
[2211.86 --> 2215.54] heard about um from you so why don't you talk a little bit about ronin and what it is and what's |
[2215.54 --> 2221.84] the purpose oh yeah so uh back in the day i was working in a computer security research group uh we had |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.