Is anyone here using it? I've only recently heard about this.
https://github.com/atlas-engineer/next
Reminds of luakit, but with Lisp.
>>1
I haven't used it either but I remember seeing it posted on /tech/ awhile back, it's good to see they're sticking with it. I do wonder how this compares to EmacsXWidgets + WebKit (the emacs package) though, it seems like a more integrated environment with even more power.
Vip
i bookmarked that link to look @ later
installed it from the aur and started using it yesterday
literally crashes all the time and when it doesn't its just slow
such a shame, it uses lisp, is keyboard driven, and has emacs bindings: it should be better and I hope they fix the browser
kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk
kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk
kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk
m
my installation fails with the following output:
<output>
WARNING: Setting locale failed.
Check the following variables for correct values:
LANG=en_US.UTF-8
LC_MONETARY=en_US.UTF-8
LC_NUMERIC=en_US.UTF-8
LC_TIME=en_US.UTF-8
<INFO> [21:57:55] next base.lisp (load-lisp-file form-fun-4) -
Loading configuration from /gnu/store/9l8n6sb9l12gabf27qf7bwykqiwfk0ap-sbcl-next-1.2.1-lib/.config/next/init.lisp...
<WARN> [21:57:55] next base.lisp (start) -
swank:*communication-style* is set to NIL, recommended value is :fd-handler
<INFO> [21:57:55] next port.lisp (run-program port) -
Current directory: /home/ngks/ZVV-volism-local/src/SchemeBBS/
<INFO> [21:57:55] next port.lisp (run-program port) -
Platform port path: /gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit
<INFO> [21:57:55] next port.lisp (run-program port) -
Platform port arguments: (--port 8082 --core-socket
http://localhost:8081/RPC2)
<INFO> [21:57:55] next port.lisp (run-program port) -
Platform port log file: /home/ngks/.local/share/next/gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit.log
Unhandled SIMPLE-ERROR in thread #<SB-THREAD:THREAD "main thread" RUNNING
{10005D05B3}>:
Couldn't execute "/gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit": No such file or directory
Backtrace for: #<SB-THREAD:THREAD "main thread" RUNNING {10005D05B3}>
0: (SB-DEBUG::DEBUGGER-DISABLED-HOOK #<SIMPLE-ERROR "Couldn't execute ~S: ~A" {100475CDE3}> #<unused argument> :QUIT T)
1: (SB-DEBUG::RUN-HOOK *INVOKE-DEBUGGER-HOOK* #<SIMPLE-ERROR "Couldn't execute ~S: ~A" {100475CDE3}>)
2: (INVOKE-DEBUGGER #<SIMPLE-ERROR "Couldn't execute ~S: ~A" {100475CDE3}>)
3: (UIOP/IMAGE:HANDLE-FATAL-CONDITION #<SIMPLE-ERROR "Couldn't execute ~S: ~A" {100475CDE3}>)
4: (SB-KERNEL::%SIGNAL #<SIMPLE-ERROR "Couldn't execute ~S: ~A" {100475CDE3}>)
5: (ERROR "Couldn't execute ~S: ~A" "/gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit" "No such file or directory")
6: (RUN-PROGRAM "/gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit" ("--port" "8082" "--core-socket" "http://localhost:8081/RPC2") :ENV NIL :ENVIRONMENT NIL :WAIT NIL :SEARCH T :PTY NIL :INPUT NIL :IF-INPUT-DOES-NOT-EXIST :ERROR :OUTPUT #P"/home/ngks/.local/share/next/gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit.log" :IF-OUTPUT-EXISTS :APPEND :ERROR :OUTPUT :IF-ERROR-EXISTS :APPEND :STATUS-HOOK NIL :EXTERNAL-FORMAT :UTF-8 :DIRECTORY NIL)
7: (UIOP/LAUNCH-PROGRAM:LAUNCH-PROGRAM ("/gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit" "--port" "8082" "--core-socket" "http://localhost:8081/RPC2") :OUTPUT #P"/home/ngks/.local/share/next/gnu/store/sgbfnhg5662by6pd142qaf2x2kib5yq1-next-gtk-webkit-1.2.1/bin/next-gtk-webkit.log" :ERROR-OUTPUT :OUTPUT)
8: ((:METHOD NEXT::RUN-PROGRAM (NEXT::PORT)) #<NEXT::PORT {1003CE3EA3}>) [fast-method]
9: (NEXT:START :WITH-PLATFORM-PORT-P T)
10: (NEXT:START-WITH-PORT)
11: (NEXT-EXEC:MAIN)
12: ((LAMBDA NIL :IN UIOP/IMAGE:RESTORE-IMAGE))
13: (UIOP/IMAGE:CALL-WITH-FATAL-CONDITION-HANDLER #<CLOSURE (LAMBDA NIL :IN UIOP/IMAGE:RESTORE-IMAGE) {10038507FB}>)
14: ((FLET SB-UNIX::BODY :IN SAVE-LISP-AND-DIE))
15: ((FLET "WITHOUT-INTERRUPTS-BODY-7" :IN SAVE-LISP-AND-DIE))
16: ((LABELS SB-IMPL::RESTART-LISP :IN SAVE-LISP-AND-DIE))
unhandled condition in --disable-debugger mode, quitting
</output>
guess it is badly packaged in guix?
>>9
https://github.com/atlas-engineer/next/tree/master/documents#user-content-run
https://github.com/atlas-engineer/next/issues/160
May 8, 2019
This should be fixed in Next 1.2.1. Feel free to re-open if you still experience the issue.
I'm running stumpwm and next browser on debian.
next browser provides the keybinding C-w for COPY-URL
but after issuing this command I can't find the url using xclip, nor can I paste it into emacs (or "yank" if you want to be more correct) -- am I pointed to the wrong X selection or something?
This is a boneheaded question I know, but emacs has been so good at handling X copy paste behavior that I haven't needed to address an issue like this in some years.
I would also be happy to hear about addressing this minor chore with a giant hammer such as org-capture, slime, etc.
(cross posted at http://textboard.org/prog/54, after feedback from some user I now realize this is a better place for the question)
I'm very interested in next and plan to use it as my primary browser once it gets a bit more mature.
Well, I tried it out after reading this thread and some docs on their site. Really interesting browser, I'm posting from it right now. I would definitely consider contributing to it in the future once I am more knowledgeable.
Right now I'm running Emacs with EXWM, and Next is a perfect fit for it.
My main concerns are security related, I'm not sure if any sophisticated script blockers have been developed for it, and it seems unlikely that support for webextensions will come any time soon (if ever), as the standard itself seems at least partially antithetical to the design of the browser.
As is made obvious by panopticlick.eff.org, by default the browser has poor tracking protection, and so in the meantime I generally won't be using it to post on sites that use JS. That limits me to messageboards such as this one (which might not be too bad, all things considered).
>>11
Sorry, I doubt that I can be of much real help here.
I would suggest going into the code of copy-url and seeing what mechanism is used to get the text, then writing your own function that gets the text using that method and piping it to xclip or something.
I've done something quite similar to that with xsel recently, and it worked pretty well as a means to get text from a mix of both emacs buffers and exwm buffers. From what I saw in the Next documentation it should be possible to do the same.
I just used an async-shell-command elisp function, and I'm sure Common Lisp would have an equivalent that probably works better, too.
>>13 here
I've been using the browser some more. It's actually quite nice. The vi-keybindings work really well, and it's easy to integrate with other programs and change the keybinds to prevent conflicts.
You can disable JS entirely using M-x noscript-mode.
I've also found that the browser works just fine over Tor, and can even be launched through torify.
are you still using it, >>15 ?
I don't have a vim background but I find the interface more appealing than qutebrowser.
There was a lightning talk about it at FOSDEM: https://fosdem.org/2020/schedule/event/next_web_browser/
Has anybody gone full "Guix System + StumpWM + Emacs + Next Browser" yet?
>>19
StumpWM is more rare these days because of EXWM.
>>18
I'm waiting for Next to reach stable on Nix.
hi everyone
>>19
I've never heard of it, thanks.
>>19
I'm not too sure, since StumpWM is still suposed to be more stable than EXWM, and cannot be effected by Emacs blocking.
>>23
Nope, EXWM is far more stable than StumpWM these days. Shit it can't even load truetype fonts without slowing down the whole damn UI. Emacs blockage is rare to non-existent if you know how to avoid it, users don't have much trouble there. BUT the real gem in using both of these WMs is the builtin support for the so called "simulation keys" (StumpWM acquired this functionality after EXWM) and this stuff makes the nEXT browser keybinding scheme a bit redundant (it still gets points for extensibility I guess). With simulation keys you can configure keybinding translations per window class, let's say make C-k send Ctrl-x in Firefox or C-x C-c to send Ctrl-q etc.
>>18
The vi shortcurts are too deeply ingrained in me. I have never found a good replacement for keyboard navigation add-on Pentadactyl. All the clones are inferior. I have to use Palemoon now, since Pentadactyl development has stalled and it cannot be ported to newer versions of Firefox.
I also have many handy commands that I'd have to rewrite. For instance, if I want to fetch a 1995 web page that is no longer available, I just type ":ar<tab>" (autocompletes to :archive) and the archive.org for this page is loaded. It's a single line in .pentadactylrc
command! archive js dactyl.open("https://web.archive.org/web/*/" + content.location.href)Is anyone using this? It's kind of interesting, but the meat is still webkit, and the web is burdensome to use anyway...
Not sure but I think I might be OP and I am now using it daily with Guix and EXWM and it's pretty great. I don't think Nyxt browser (new name) is made useless by EXWM simulation keys since you can do things like execute arbitrary js in the browser via parenscript. I'm not aware of a way to do that with just EXWM. Is it a good idea? Probably not I guess.
I can't offer any encouraging words as far as security -- I definitely don't use it for anything risky like banking etc.
I agree that we still have not recreated the good old days of vimperator/pentadactyl but Nyxt makes me more optimistic.
Web does suck though.
>>27
Pentadactyl still works fine with Palemoon.