A place to cache linked articles (think custom and personal wayback machine)
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

7 月之前
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105
  1. title: Redis Renamed to Redict
  2. url: https://andrewkelley.me/post/redis-renamed-to-redict.html
  3. hash_url: d58c7619eec894c5c069244114ea1df5
  4. archive_date: 2024-03-23
  5. og_image: https://andrewkelley.me/img/redis-screenshot-light.png
  6. description: In other words, Redict is the true spiritual successor to what was once Redis.
  7. favicon:
  8. language: en_US
  9. <p><a href="https://redict.io/">Redict</a> was originally
  10. created by Salvatore Sanfilippo under the name "Redis".
  11. Around 2018 he started losing interest in the project to pursue a science fiction career and gave
  12. stewardship of the project to <a href="https://en.wikipedia.org/wiki/Redis_(company)">Redis Labs</a>.</p>
  13. <p>
  14. I think that was an unfortunate move because their goal is mainly to extract profit from
  15. the software project rather than to uphold the ideals of Free and Open Source Software.
  16. On March 20, 2024, they
  17. <a href="https://github.com/redis/redis/pull/13157">changed the license to be proprietary</a>,
  18. a widely unpopular action. You know someone is up to no good when they write
  19. "Live long and prosper 🖖" directly above a meme of Darth Vader.
  20. </p>
  21. <img class="light" alt="screenshot of the PR merging with many downvotes and darth vader" src="https://andrewkelley.me/img/redis-screenshot-light.png">
  22. <h2 id="license-problems">What are the actual problems with these licenses?</h2>
  23. <p>In short summary, the licenses limit the freedoms of what one can do with
  24. the software in order for Redis Labs to be solely enriched, while asking for
  25. volunteer labor, and having <a href="https://github.com/redis/redis/pull/13157#issuecomment-2014737480">already
  26. benefitted from volunteer labor</a>, that is and was <a href="https://news.ycombinator.com/item?id=39775468">generally offered only
  27. because it enriches <em>everybody</em></a>.</p>
  28. <p><a href="https://ssplisbad.com/">SSPL is BAD</a></p>
  29. <h2 id="legality">Are they allowed to do this?</h2>
  30. <p>All the code before the license change is available under the previous license (BSD-3),
  31. however it is perfectly legal to make further changes to the project under a different license.</p>
  32. <p>This means that it is also legal to <em>fork</em> the project from before the license change,
  33. and continue maintaining the project without the proprietary license change. The only problem
  34. there would be, the project would be missing out on all those juicy future
  35. contributions from Redis Labs... wait a minute, isn't the project already <em>done</em>?
  36. </p>
  37. <h2 id="already-completed">Redict is a Finished Product</h2>
  38. <p>Redict already works great. Lots of companies already use it in production and have
  39. been doing so for many years.</p>
  40. <p>In <a href="https://andrewkelley.me/post/why-we-cant-have-nice-software.html">Why We Can't Have Nice Software</a>,
  41. I point out this pattern of needless software churn in the mindless quest for
  42. profit. This is a perfect example occurring right now. Redict has already reached its
  43. peak; it does not need any more serious software development to occur.
  44. It does not need to <a href="https://redis.com/blog/the-future-of-redis/">pivot
  45. to AI</a>. It can be maintained for decades to come with minimal effort. It can
  46. continue to provide a high amount of value for a low amount of labor. That's
  47. the entire point of software!</p>
  48. <p><strong>Redict does not have any profit left to offer</strong>. It no longer
  49. needs a fund-raising entity behind it anymore. It just needs a good project steward.</p>
  50. <h2 id="drew">Drew DeVault is a Good Steward</h2>
  51. <p><a href="https://drewdevault.com/">Drew</a> is a controversial person, I
  52. think for two reasons.</p>
  53. <p>One, is that he has a record of being rude to many people in the past -
  54. including myself. However, in a <a href="https://lore.kernel.org/lkml/CA+55aFy+Hv9O5citAawS+mVZO+ywCKd9NQ2wxUmGsz9ZJzqgJQ@mail.gmail.com/">similar manner as Linus Torvalds</a>,
  55. Drew has expressed what I can only interpret as sincere regret for such interactions, as
  56. well as a pattern of improved behavior. I was poking through his blog to try to find
  57. example posts of what I mean, and it's difficult to pick them out because he's such a
  58. prolific writer, but perhaps <a href="https://drewdevault.com/2022/05/30/bleh.html">this one</a>
  59. or maybe <a href="https://drewdevault.com/2023/05/01/2023-05-01-Burnout.html">this one</a>.
  60. I'm a strong believer in applying
  61. <a href="https://en.wikipedia.org/wiki/Tit_for_tat">the best game theory strategy</a>
  62. to society: people should have consequences for the harm that they do, but then they should
  63. get a chance to start cooperating again. I can certainly think of "cancelable" things
  64. I have done in the past, that I am thankful are not public, and I cringe every
  65. time I remember them.
  66. </p>
  67. <p>Secondly, and I think this is actually the more important point, Drew has been
  68. an uncompromising advocate of Free and Open Source Software his entire life,
  69. walking the walk more than anyone else I can think of. It's crystal clear that
  70. this is the driving force of his core ideology that determines all of his decision
  71. making. He doesn't budge on any of these principles and it creates conflicts
  72. with people who are trying to exploit FOSS for their own gains. For example, when
  73. you <a href="https://drewdevault.com/2023/07/04/Dont-sign-a-CLA-2.html">call out SourceGraph</a>
  74. you basically piss off everyone who has SourceGraph stock. Do enough of these callouts,
  75. and you've pissed off enough people that there's an entire meme subculture around hating you.</p>
  76. <p>Meanwhile, Drew maintains <a href="https://gitlab.freedesktop.org/wlroots/wlroots">wlroots</a> and <a href="https://swaywm.org/">Sway</a>, and runs <a href="https://sourcehut.org/">a sustainable business</a> on top of
  77. <a href="https://sr.ht/~sircmpwn/sourcehut/">Free and Open Source Software</a>.
  78. SourceHut has a dependency on Redict, so it naturally follows that Drew wants
  79. to keep his supply chain FOSS.</p>
  80. <h2 id="redis-is-the-fork">Redis is the Fork</h2>
  81. <p>
  82. The only thing Redis has going for it, as a software project, is the brand name.
  83. Salvatore is long gone. The active contributors who are working on it are, like I said,
  84. pivoting to AI. Seriously, here's a quote from <a href="https://redis.com/blog/the-future-of-redis/">The Future of Redis</a>:
  85. </p><blockquote>
  86. <p>Making Redis the Go-To for Generative AI</p>
  87. <p>we’re staying at the forefront of the GenAI wave</p>
  88. </blockquote>
  89. <p>
  90. Meanwhile, Redict has an actual Free and Open Source Software movement behind
  91. it, spearheaded by Drew DeVault, who has a track record of effective open
  92. source project management.
  93. </p>
  94. <p>
  95. In other words, Redict is the true spiritual successor to what was once Redis.
  96. The title of this blog post is not spicy or edgy; it reflects reality.
  97. </p>