๐พ Archived View for gemi.dev โบ gemini-mailing-list โบ 000113.gmi captured on 2023-12-28 at 15:41:27. Gemini links have been rewritten to link to archived content
โฌ ๏ธ Previous capture (2023-11-04)
-=-=-=-=-=-=-
Hey all! I was working on this for the last weekend or so, and I believe it's now ready enough to share. Announcing: Moonlander[1], the fanciest graphical Gemini client in the entire solar system [citation needed]. [1]: https://sr.ht/~admicos/moonlander A screenshot: https://files.ecmelberk.com/img/1589891854.png Please note that it's still under development, and there are a lot of known bugs. Still, why not take it for a spin? The rest of this mail will be an excerpt from the readme file: ---- ## Features - Custom, themeable rendering engine via Cairo & Pango - Tries to follow Gnome HIG ### Known Bugs - TLS Certificates aren't verified - Somewhat high resource usage (for a Gemini client) - No cross-protocol linking (yet) - Cannot navigate backwards through redirections - Renderer doesn't behave "native" - Cannot select/copy text - No interaction other than mouse clicks on links and scrolling ### Planned Features - Tabs - Render more than just text/gemini and plaintext. - Planned: Markdown & images - Possibly support other protocols - Gopher, etc. - Definitely not HTTP, unless excluding HTML - Syntax highlighting (?) - Waiting on text/gemini preformatting annotations to be somewhat standardized
On Tuesday 19 May 2020 14:43, Ecmel Berk Canl?er <me at ecmelberk.com> wrote: > - Render more than just text/gemini and plaintext. > - Planned: Markdown & images A Markdown Browser... what a nice idea!
Wow, that screenshot is looking good! I'm excited for the planned markdown and image rendering. Feel free to send an update to us as more features are added. Also come join us on IRC: #gemini at tilde.chat (irc.tilde.chat is the actual server). I will compile and try it out! makeworld ??????? Original Message ??????? On Tuesday, May 19, 2020 8:43 AM, Ecmel Berk Canl?er <me at ecmelberk.com> wrote: > Hey all! > > I was working on this for the last weekend or so, and I believe it's > now ready enough to share. > > Announcing: Moonlander[1], the fanciest graphical Gemini client in the > entire solar system [citation needed]. > > [1]: https://sr.ht/~admicos/moonlander > > A screenshot: https://files.ecmelberk.com/img/1589891854.png > > Please note that it's still under development, and there are a lot of > known bugs. Still, why not take it for a spin? > > The rest of this mail will be an excerpt from the readme file: > > ------------------------------------------------------------------------- --------------------------------------------------------------------------- --------------------------------------------------------------------------- --------------------------------------------------------------------------- --------------------------------------------------------------------------- --------------------------------------------------------------------------- ----------------------------------------------------- > > Features > > --------- > > - Custom, themeable rendering engine via Cairo & Pango > - Tries to follow Gnome HIG > > ### Known Bugs > > - TLS Certificates aren't verified > - Somewhat high resource usage (for a Gemini client) > - No cross-protocol linking (yet) > - Cannot navigate backwards through redirections > - Renderer doesn't behave "native" > - Cannot select/copy text > - No interaction other than mouse clicks on links and scrolling > > ### Planned Features > > - Tabs > - Render more than just text/gemini and plaintext. > - Planned: Markdown & images > - Possibly support other protocols > - Gopher, etc. > - Definitely not HTTP, unless excluding HTML > - Syntax highlighting (?) > - Waiting on text/gemini preformatting annotations to be somewhat > standardized >
On Tue, 19 May 2020 20:52:02 +0000 colecmac at protonmail.com wrote: > Wow, that screenshot is looking good! I'm excited for the planned > markdown and image rendering. Thanks :) > Feel free to send an update to us as more features are added. Will do! (Assuming I don't get bored and stop working on it) > Also come join us on IRC: #gemini at tilde.chat (irc.tilde.chat is > the actual server). Already there (admicos)
I just wrote about why Moonlander hasn't seen any commits after it's
first 3 days. Might be interesting to some of you.
gemini://ebc.li/posts/how-hard-can-it-be.gmi
tldr: Text is hard and Moonlander _might_ get abandoned.
ps: Not entirely sure on mailing list etiquette, so I made this a reply
of the original announcement. Hope that's fine!
pps: Funnily enough, Moonlander might not be able to load that page. At
least it didn't on my system. Probably some incompatibility with the
new server.
On 04-Aug-2020 20:09, Ecmel Berk Canl?er wrote: > I just wrote about why Moonlander hasn't seen any commits after it's > first 3 days. Might be interesting to some of you. > > gemini://ebc.li/posts/how-hard-can-it-be.gmi That's a great post - thanks for the write up. Yes text rendering is really hard when you go beyond the basics, so I agree it saves a lot of frustration to use an existing library if you can. If it is just text selection that is holding you back and you want to continue with your UI experiments, why not abandon text selection? If the user really wants text selection, you could offer a view-source of the raw content in a vanilla text window, and the user can select from there in a very straightforward way. You could even scroll to the line offset to help if you wanted. Best Wishes - Luke
> If it is just text selection that is holding you back and you want to > continue with your UI experiments, why not abandon text selection? If > the user really wants text selection, you could offer a view-source > of the raw content in a vanilla text window, and the user can select > from there in a very straightforward way. You could even scroll to > the line offset to help if you wanted. Aside from being an interesting experiment, I want Moonlander to be something polished. Having to open secondary windows to do something as "basic" (in terms of user interaction) as copy/paste wouldn't be good design. Aside from that, the infrastructure for at least basic text selection would be required to handle inline links in formats like Markdown, if they were to get implemented.
---