home.social

Search

481 results for “develwithoutacause”

  1. Playing a bit with #Preact / #JSX today and am feeling the pain of not having host elements. I've realized that because a "component" is really just a #VDOM template, there is no implicit contract around what it actually rendered.

    As an arbitrary example, what if I want to set a style on a `<Foo />` component? What selector should I use for that style? I have no idea what `<Foo />` will actually generate, or even how many top-level elements it will output. The `Foo` component _must_ expose some kind of `className` or other API to provide this behavior, it's not implicit at all.

    Maybe this is good because it provides stronger encapsulation. But it can also lead to bugs. If I render `<Header />` a `header` CSS selector is _probably_ right. But what if `<Header />` actually renders:

    <div>
    <header>...</header>
    </div>

    Then a `header` selector isn't actually the top-level element and my style could easily be wrong.

    Just some idle thoughts about the lack of host elements.

  2. Playing a bit with #Preact / #JSX today and am feeling the pain of not having host elements. I've realized that because a "component" is really just a #VDOM template, there is no implicit contract around what it actually rendered.

    As an arbitrary example, what if I want to set a style on a `<Foo />` component? What selector should I use for that style? I have no idea what `<Foo />` will actually generate, or even how many top-level elements it will output. The `Foo` component _must_ expose some kind of `className` or other API to provide this behavior, it's not implicit at all.

    Maybe this is good because it provides stronger encapsulation. But it can also lead to bugs. If I render `<Header />` a `header` CSS selector is _probably_ right. But what if `<Header />` actually renders:

    <div>
    <header>...</header>
    </div>

    Then a `header` selector isn't actually the top-level element and my style could easily be wrong.

    Just some idle thoughts about the lack of host elements.

  3. Playing a bit with #Preact / #JSX today and am feeling the pain of not having host elements. I've realized that because a "component" is really just a #VDOM template, there is no implicit contract around what it actually rendered.

    As an arbitrary example, what if I want to set a style on a `<Foo />` component? What selector should I use for that style? I have no idea what `<Foo />` will actually generate, or even how many top-level elements it will output. The `Foo` component _must_ expose some kind of `className` or other API to provide this behavior, it's not implicit at all.

    Maybe this is good because it provides stronger encapsulation. But it can also lead to bugs. If I render `<Header />` a `header` CSS selector is _probably_ right. But what if `<Header />` actually renders:

    <div>
    <header>...</header>
    </div>

    Then a `header` selector isn't actually the top-level element and my style could easily be wrong.

    Just some idle thoughts about the lack of host elements.

  4. Playing a bit with #Preact / #JSX today and am feeling the pain of not having host elements. I've realized that because a "component" is really just a #VDOM template, there is no implicit contract around what it actually rendered.

    As an arbitrary example, what if I want to set a style on a `<Foo />` component? What selector should I use for that style? I have no idea what `<Foo />` will actually generate, or even how many top-level elements it will output. The `Foo` component _must_ expose some kind of `className` or other API to provide this behavior, it's not implicit at all.

    Maybe this is good because it provides stronger encapsulation. But it can also lead to bugs. If I render `<Header />` a `header` CSS selector is _probably_ right. But what if `<Header />` actually renders:

    <div>
    <header>...</header>
    </div>

    Then a `header` selector isn't actually the top-level element and my style could easily be wrong.

    Just some idle thoughts about the lack of host elements.

  5. Playing a bit with / today and am feeling the pain of not having host elements. I've realized that because a "component" is really just a template, there is no implicit contract around what it actually rendered.

    As an arbitrary example, what if I want to set a style on a `<Foo />` component? What selector should I use for that style? I have no idea what `<Foo />` will actually generate, or even how many top-level elements it will output. The `Foo` component _must_ expose some kind of `className` or other API to provide this behavior, it's not implicit at all.

    Maybe this is good because it provides stronger encapsulation. But it can also lead to bugs. If I render `<Header />` a `header` CSS selector is _probably_ right. But what if `<Header />` actually renders:

    <div>
    <header>...</header>
    </div>

    Then a `header` selector isn't actually the top-level element and my style could easily be wrong.

    Just some idle thoughts about the lack of host elements.

  6. This change also has a side benefit of giving you full control over the `ts_project` which compiles your prerendering logic. No more awkward wrapping in #rules_prerender, much cleaner!

  7. This change also has a side benefit of giving you full control over the `ts_project` which compiles your prerendering logic. No more awkward wrapping in #rules_prerender, much cleaner!

  8. This change also has a side benefit of giving you full control over the `ts_project` which compiles your prerendering logic. No more awkward wrapping in #rules_prerender, much cleaner!

  9. This change also has a side benefit of giving you full control over the `ts_project` which compiles your prerendering logic. No more awkward wrapping in #rules_prerender, much cleaner!

  10. This change also has a side benefit of giving you full control over the `ts_project` which compiles your prerendering logic. No more awkward wrapping in , much cleaner!

  11. Just landed a major update to #rules_prerender yesterday. I redesigned the core component API to better express dependencies between different "slices" of a component. For example, if you only depend on the client-side JavaScript, the bundler will still find the associated CSS styles.

    github.com/dgp1130/rules_prere

    Really excited about the direction here! Next step is the actual documentation site.

  12. Just landed a major update to #rules_prerender yesterday. I redesigned the core component API to better express dependencies between different "slices" of a component. For example, if you only depend on the client-side JavaScript, the bundler will still find the associated CSS styles.

    github.com/dgp1130/rules_prere

    Really excited about the direction here! Next step is the actual documentation site.

  13. Just landed a major update to #rules_prerender yesterday. I redesigned the core component API to better express dependencies between different "slices" of a component. For example, if you only depend on the client-side JavaScript, the bundler will still find the associated CSS styles.

    github.com/dgp1130/rules_prere

    Really excited about the direction here! Next step is the actual documentation site.

  14. Just landed a major update to #rules_prerender yesterday. I redesigned the core component API to better express dependencies between different "slices" of a component. For example, if you only depend on the client-side JavaScript, the bundler will still find the associated CSS styles.

    github.com/dgp1130/rules_prere

    Really excited about the direction here! Next step is the actual documentation site.

  15. Just landed a major update to yesterday. I redesigned the core component API to better express dependencies between different "slices" of a component. For example, if you only depend on the client-side JavaScript, the bundler will still find the associated CSS styles.

    github.com/dgp1130/rules_prere

    Really excited about the direction here! Next step is the actual documentation site.

  16. @slightlyoff I use #JSON5 in a Chrome Extension I own solely to comment the manifest and it works great.

    I do think the bigger issue is that JSON is a terrible configuration language (but great wire format!). We should be using #TOML, #YAML, or something else more targeted for that use case.

  17. @slightlyoff I use #JSON5 in a Chrome Extension I own solely to comment the manifest and it works great.

    I do think the bigger issue is that JSON is a terrible configuration language (but great wire format!). We should be using #TOML, #YAML, or something else more targeted for that use case.

  18. @slightlyoff I use #JSON5 in a Chrome Extension I own solely to comment the manifest and it works great.

    I do think the bigger issue is that JSON is a terrible configuration language (but great wire format!). We should be using #TOML, #YAML, or something else more targeted for that use case.

  19. @slightlyoff I use #JSON5 in a Chrome Extension I own solely to comment the manifest and it works great.

    I do think the bigger issue is that JSON is a terrible configuration language (but great wire format!). We should be using #TOML, #YAML, or something else more targeted for that use case.

  20. @slightlyoff I use in a Chrome Extension I own solely to comment the manifest and it works great.

    I do think the bigger issue is that JSON is a terrible configuration language (but great wire format!). We should be using , , or something else more targeted for that use case.

  21. A sincere "fuck you" to #GetYourGuide (and any other app) which listed a tour fully online, let me pay and complete the transaction, and email me the result. But instead of emailing me a QR code, it emails me a link to their *native app*, forces me to install the 40 MB binary on mobile roaming data, remember the password to log in to the same one-off account made to purchase the ticket, *and then* show me the QR code to show the tour guide. Just email me the damn QR code!

    This accomplishes *nothing* but adding 14 clicks, 5 - 20 minutes to the check in time (depending on how easily the user guesses their password), and requiring a mobile data connection.

  22. As for what I hope would be next (keep in mind I did not play #RE2Remake or #RE3Remake): I don't think it's worth remaking #RE5 given it's general reception. I'd much rather see this team make a brand new Resident Evil game in this engine. They've clearly got a solid track record at this point and we never really got a "good" over-the-shoulder RE game after 4 (until the remakes).

    I'd love to see an entirely new, from scratch experience built with these mechanics and engine as a starting point.

    This does kind of put the team in the same position as the RE franchise was post-RE4, where they previously fucked it all up. However, RE4 was a very unique "lightning in a bottle" moment where it got so many things right on the first try. Now that we've had several games of bad ideas and the RE 2/3 remakes, I feel like we should have a better sense of what does and doesn't work which will hopefully lead to a better end product than RE5 was. Just what I'd like to see anyways.

  23. I want a #SuperSmashBros-clone where all the characters are mascots from children's cereal.

    Imagine a battle royal between the #LuckyCharms leprechaun, #ToucanSam, the #Trix rabbit, and #TonyTheTiger.

  24. Speaking of not writing documentation: I also prototyped a very cool idea for how to greatly simplify the developer experience of #rules_prerender *and* make it understand even more complex dependency graphs.

    Still more experimenting to be done, but I'm hopeful this is a viable strategy and will be a big improvement for developers! 😁

  25. Speaking of not writing documentation: I also prototyped a very cool idea for how to greatly simplify the developer experience of #rules_prerender *and* make it understand even more complex dependency graphs.

    Still more experimenting to be done, but I'm hopeful this is a viable strategy and will be a big improvement for developers! 😁

  26. Speaking of not writing documentation: I also prototyped a very cool idea for how to greatly simplify the developer experience of #rules_prerender *and* make it understand even more complex dependency graphs.

    Still more experimenting to be done, but I'm hopeful this is a viable strategy and will be a big improvement for developers! 😁

  27. Speaking of not writing documentation: I also prototyped a very cool idea for how to greatly simplify the developer experience of #rules_prerender *and* make it understand even more complex dependency graphs.

    Still more experimenting to be done, but I'm hopeful this is a viable strategy and will be a big improvement for developers! 😁