60 replies

1. Dave 🧱 5 min later

   @zachleat 👀

   1. Khalid ⚡ 10 min later

      @DavidDarnes @zachleat 👀 👀 👀

2. Andy Bell 10 min later

   @zachleat 👂⬅️

3. elle mundy 11 min later

   @zachleat 😯

4. Paul Everitt 1 hr later

   @zachleat My now-abandoned Gatsby GraphQL schema extensions are rolling over in their graves.

5. Joseph Kohlmann 6 hr later

   @zachleat Do I even *wanna* know? 🫠🙈

   1. Zach Leatherman 6 hr later

      @jkohlmann if you use Gatsby, probably 😅

      1. Joseph Kohlmann 7 hr later

         @zachleat Thankfully I'm Gatsby-free over here 🙃

6. Zach Leatherman 4 days later

   Looks like the cat is out of the bag now, but still no official word from the company: https://twitter.com/FredKSchott/status/1693007599803752638

   The top contributor to Gatsby: https://twitter.com/lekoarts_de/status/1685942410776342528

   Another Gatsby employee: https://twitter.com/wardpeet/status/1693014604694061194

   Saw also a public Slack post from the founder saying Gatsby was in “maintenance mode.”

   1. Brian Rinaldi 4 days later

      @zachleat I'd heard the same from my sources. Post-acquisition, I kept telling folks that my expectation was that they'd "donate" it to the OpenJS Foundation or something like that to "give it back to the community" (i.e. we don't maintain it anymore).

      That could still happen but the current trajectory looks worse than my prediction.

      1. Zach Leatherman 4 days later

         @remotesynth I appreciate that tough choices are being made but the lack of communication is not a good look

         1. Brian Rinaldi 4 days later

            @zachleat Agreed. I know folks on Gatsby Cloud and they keep being told that the features will be rolled into standard Netlify, but I don't see how that aligns with the current plans for putting Gatsby in maintenance mode. I can't see them putting engineering resources on an EOL project.

            1. Zach Leatherman 4 days later

               @remotesynth I don’t think I’m allowed to say but I did play darts last weekend and got a bullseye

   2. Feoh 4 days later

      @zachleat I really hate to say this, because I'm acquainted with some of the people involved, but GatsbyJS was/is a tire fire to actually work with.

      I wrote a blog post on my experience, and it wasn't pretty. I tried like heck to be as open minded and fair as I could.

      https://www.feoh.org/posts/mis-adventures-in-gatsbyjs

      #javascript #gatsby #webdev

      1. Zach Leatherman 4 days later

         @feoh yeah the merits of the framework are probably a moot point now

         1. Feoh 4 days later

            @zachleat I dunno. I think they still have a place. I'm building a @djangonews app for myself right now as I need a CRUD thing and I don't want to have to spend months learning something super heavy or alternatively so light I have to build everything from scratch.

            1. Zach Leatherman 4 days later

               @feoh ah, to be clear—specifically I don’t believe the merits of Gatsby are relevant if they are retiring the thing

   3. Zach Leatherman 4 days later

      Perhaps controversially—I believe the same could very easily happen to any VC-funded framework.

      What happens when you have multiple FTE on payroll and you exhaust your runway without a viable strategy to monetize?

      1. Steve Workman 4 days later

         @zachleat "community stewardship" - always the ideal but almost impossible to find.

         I feel for them, and hope they all land on their feet

      2. Jared White (ResistanceNet ✊) 4 days later

         @zachleat I don't see what's controversial about that. In fact, obvious, no? 😅

         1. Rijk 🐰 ✨ 4 days later

            @jaredwhite @zachleat Doesn't the same apply to any project no matter where the funding is coming from? 🤔

            1. Zach Leatherman 4 days later

               @rijk @jaredwhite not really, no. They don’t just have to figure out monetization/profitability—they need to figure out hyper-monetization 😬

               1. Rijk 🐰 ✨ 4 days later

                  @zachleat @jaredwhite ah yeah if the goal is growth and profit no matter what then absolutely!

               2. Paul Everitt 5 days later

                  @zachleat @rijk @jaredwhite For those looking for the answer…this is it.

                  VCs expect 19 of 20 fundings to fail. Number 20 has to be a Google. You can't just succeed and have a nice steady customer base.

                  1. Jared White (ResistanceNet ✊) 5 days later

                     @pauleveritt @zachleat @rijk Exactly. The VC model quite honestly is based on "outrageous success" or undeniable failure. Plodding along, business as usual—not OK at all. Arguably the worst outcome in fact.

      3. Zach Leatherman 4 days later

         “We use React. Its power seems inescapable. So did the divine right of kings.”—Ursula K. Le Guin, probably

         1. Khalid ⚡ 4 days later

            @zachleat

         2. Private reply 5 days later
            1. Private reply 5 days later
         3. woolie 6 days later

            @zachleat I've managed to have a career without writing a single line of react based code.

      4. Brian LeRoux 💚 4 days later

         @zachleat not just can. Likely will. Foundations and the concepts of open governance exist to fix real problems but we are enjoying a period of relearning the hard way for some fucked reason I can't decrypt.

         1. Christopher “boneskull” Hiller 5 days later

            @brianleroux @zachleat OSS is treated as marketing, and marketing works

            1. Brian LeRoux 💚 5 days later

               @boneskull @zachleat wonder how folks who deployed it to production workloads feel. Probably not super great. (Assuming it has any production workloads..)

               1. Christopher “boneskull” Hiller 5 days later

                  @brianleroux @zachleat they had it coming

                  1. Christopher “boneskull” Hiller 5 days later

                     @brianleroux @zachleat i know one thing though: it’ll get worse before it gets better. if it gets better

                     1. Brian LeRoux 💚 5 days later

                        @boneskull @zachleat we all know the same fate in store for NextJS

                        1. Zach Leatherman 5 days later

                           @brianleroux @boneskull whether it happens pre-IPO or post-IPO is the only question 🫣

                           1. Brian LeRoux 💚 5 days later

                              @zachleat @boneskull

                        2. Christopher “boneskull” Hiller 5 days later

                           @brianleroux @zachleat like, it’s one thing if you buy some SaaS subscription that goes ass up. but once you build your shit on top of a framework, you’re hosed.

                           1. Christopher “boneskull” Hiller 5 days later

                              @brianleroux @zachleat your saving grace is that there might be enough dead frogs in the pot with you to fork into a community project

                              1. Christopher “boneskull” Hiller 5 days later

                                 @brianleroux @zachleat how many of these projects remain ongoing concerns is unknown to me. it’s there in theory

                                 1. Brian LeRoux 💚 5 days later

                                    @boneskull @zachleat personally think we explored transpiling dialects to the logical extremes and are only now in the "finding out" phase of how much of an evolutionary dead end that approach is.

                                    1. Corbin 5 days later

                                       @brianleroux @boneskull @zachleat This is my favorite takeaway. The reason to target JS was that it was "close to the metal" in terms of portable Web media; the entire cottage industry of "transpiler to JS" is built upon that requirement. I can't help but speculate that WASM is part of the picture; the overwhelming complexity had to collapse somehow, and shifting foundations would be a reasonable cause.

                                       There was discussion about this recently, on Lobsters: https://lobste.rs/s/id9jet/transpiler_meaningless_word

                                       1. Brian LeRoux 💚 5 days later

                                          @corbin @boneskull @zachleat text to text is pretty lossy and interpreted code always on a back foot wrt optimization so wasm def seems inevitable.

                                          But I've been saying this for a decade at least now too. I have no issue with the techniques but question their assumed universal application.

                                 2. Christopher “boneskull” Hiller 5 days later

                                    @brianleroux @zachleat can’t imagine “surprise! need to commit resources to this new OSS project” goes over too well with mgmt though

                                    1. Brian LeRoux 💚 5 days later

                                       @boneskull @zachleat mgmt def starts to raise eyebrows at rewrites that don't add new value. Partially why I'm skeptical a lot of these kinds of projects end up as part of production workloads. No business wants an annual rewrite for new HTML rendering modes.

                                       1. Christopher “boneskull” Hiller 5 days later

                                          @brianleroux @zachleat I mean, somebody was using gatsby for something important. they wouldn’t have survived as long as they did without at least some customers, right?

                                          1. Brian LeRoux 💚 5 days later

                                             @boneskull @zachleat given their investors passed on us because 'github stars' I'm skeptical

                                          2. busticated 5 days later

                                             @boneskull @brianleroux @zachleat i used it for a marketing website like~5yrs ago. important? idk 🤷‍♂️ last i heard it was still chugging along tho.

                                             1. Zach Leatherman 5 days later

                                                @busticated @boneskull @brianleroux (you may want to read at the beginning of the thread)

                                                1. busticated 5 days later

                                                   @zachleat @boneskull @brianleroux oh yeah i’m aware (check yr favs 😉). not great!

               2. Zach Leatherman 5 days later

                  @brianleroux @boneskull

      5. Josh Wayne 4 days later

         @zachleat I had no idea it was a VC backed project. So glad I didn't sink too much energy into it back when I was exploring it for a project.

         1. Zach Leatherman 4 days later

            @joshwayne this post might have helped you! https://www.zachleat.com/web/monetization/

            1. Josh Wayne 5 days later

               @zachleat That certainly would have helped!

      6. Paul Everitt 5 days later

         @zachleat As someone that had a VC-backed framework in 1999 (Python's first "killer app”), I can confirm this. I spoke about this effect on a recent podcast.

         1. Zach Leatherman 5 days later

            @pauleveritt I’d love to listen—got a link?

            1. Paul Everitt 5 days later

               @zachleat Here’s the episode: https://pythonpeople.fm/episodes/paul-everitt

               7:55 I talk about foundations. 8:57 about Python outliving Guido as BDFL. 13:22 gets into the VC culture, open source, and...milking cows.

               1. Zach Leatherman 5 days later

                  @pauleveritt Thank you!!

                  1. Paul Everitt 5 days later

                     @zachleat My pleasure. Thanks for bringing this topic up.

                     BTW...@khalidabuhakmeh and I (and others) are about to deploy our 11ty+esbuild+Vite+TSX+schema validation knowledge base thingy. The Gatsby replacement. Think there would be any interest if we did an explainer? Might you join us for a webinar showing it?