How would y'all rate the overall health of the #emacs ecosystem? I'm thinking about making the switch, but I'd want to do it right and don't want to invest the time in a platform without a bright future.
StackOverflow reports that emacs fell from 5.2% to 4.1% from 2016 -2018, which seems large in percentage terms. Does it feel like usage is falling off?
And how is John Wiegley as the new maintainer? I know he had big plans about growing the core team, but I never heard how that turned out.
@codesections I would say do what makes things interesting.
@codesections I've seen several developers abandon Emacs in general use for language specific IDEs.
@codesections emacs has been around for over 40 years. It's not going anywhere any time soon.
> emacs has been around for over 40 years. It's not going anywhere any time soon.
fair enough, guess I deserve that :D
But I'm wondering less about it going somewhere and more about the momentum of the project/plugins.
I mean, I'm *happy* with vim, so if I tried out emacs, it would be for the extensibility. And that extensibility is less valuable if fewer people are making plugins with it (even if the platform doesn't go anywhere)
@mike put differently, vimscript may well be awful to write extensions in (a d it pretty much is) but if there are 1× as many people hacking away at vim plugins, it might still end up being about as extensible as emacs in practice (from a plugin-users point of view, anyway. Of course there's also the mater of your own config, but a. vimrc can go a long way)
@codesections Yea, but those numbers are results from a developer survey, and emacs is used for more than coding. Also, consider the usage numbers from their 2015 survey, which showed emacs usage at 3.8%. I'm guessing that the usage numbers you're going to see for editors like vi(m) and emacs are going to fluctuate, maybe widely over the years. Still, they're mainstays of the environment, and I think they'll be around and "healthy" for years to come.
@codesections Probably biased, but I find Emacs more appropriate for programming than vim (I use vim for single file edits).
But, you will likely need to invest some time initially to get started (same as vim).
As for the ecosystem, Emacs actually has a built-in package manager. Some notable packages: helm, magit, projectile, evil (vim mode). There's also the famous org-mode. Could be life changing :)
I don't think SO surveys are representative. I think they are more skewed towards webdev.
Going by the emacs mailing lists and blogs i follow things seem stable enough. Org-mode & magit are definitely drivers for many but if you check out MELPA you can see what packages are available.
Atom & VSE seem to be management favs in shops where your not given a choice by the pointy ear brigade.
@codesections All I know is that John is a very nice guy. I spent a weekend with him in Florence many years ago. And in all the years of online communications, he was always nice, supportive, never short tempered. Emacs is mostly growing via independent packages that are made available via its package managers. Who knows about a bright future? If other editors grow faster, Emacs loses shares and still grows. That’s why I don’t care about those numbers.
@codesections i seem to encounter no fewer (and no more) emacs devotees than i ever did - it was never in my experience that much of the population, even when i counted myself part of it.
i tried spacemacs a while ago, and though i wound up scuttling back to vim like usual, i was pretty astonished at the depth and breadth of what was available there. i assume that feeling probably extends to the emacs ecosystem as a whole.
@codesections I got switched-off emacs when trying to configure it as a productive IDE and found that you have to manually build and install some dependencies. In this day and age... really?
@codesections Considering that over the last few years new packages have been developed and widely adopted (magit, projectile, dumb-jump, evil, ivy, helm, ...) and even included into the standard distribution (org) I'd say the situation is quite ok. There's also some hope that with preconfigs like Spacemacs or Doom new users will be attracted and help developing too.
> considering that over the last few years new packages have been developed and widely adopted (magit, projectile, dumb-jump, evil, ivy, helm, ...)
Interesting! My (apparently ignorant!) impression was that those were mainstays of the ecosystem that dated back years. From a quick search, it looks like Projectile is from ~2011, which is newer than I would have guessed. And Helm is from 2015, which is *much* newer than I would have thought.
@codesections The probably newest package I use is forge (december 2018), by the magit team, to implement a uniform github/gitlab/etc. interface within magit, so there's still quite a lot going on. Other than than, old packages rarely break for no reason. Most of the time, it's external tools that it would depend on that have changed. A lot of org related packages are also pretty new.
The probably largest change for users in the last few years has been use-package, which really makes configuration a lot cleaner. See for example my config (https://zge.us.to/emacs.d.html) compared to a few of the older ones I mention.
All in all, you don't really have to worry about the ecosystem. It's stable _and_ alive, and I don't see that changing so quickly.
@codesections The introduction of elpa (and non GNU spin offs marmalade and melpa) have brought a lot of new vigor to the ecosystem. Customising your emacs is no longer a case of downloading random scraps into a gradually less maintainable heap of init.el and things like use-package have really helped with maintainable configs @zge
@codesections Your post makes me wonder what is the marketshare of Vi(m) and other editors/IDEs. Is there a new lead horse? I hate to admit it but VSCode is really nice. But so are many others.
I think you've got nothing to worry about. Emacs will be with us for a long time to come.
> Your post makes me wonder what is the marketshare of Vi(m) and other editors/IDEs.
The StackOverflow developer survey is my go-to source for data like this. It shows VSCode in the lead, used by 34.9% of developers. Vim has 25.8%, and Emacs 4.1%.
(And VSCode's growth is *amazing*. It was at 7.2% in 2016)
@codesections Wow! That is spectacular. It's very nice and deserving.
I have a story as to why but I've always tried to be editor flexible, figuring whatever machine I ended up working on, I had to be able to use what was available. I can use just about ant editor but I'm not a master of any of them.
Years ago, Brief was my go to. I still have a 3.5" floppy somewhere that I used to carry with me.
> How would y'all rate the overall health of the #emacs ecosystem?
My opinion is mildly biased, but I do see new folks joining the #Emacs community on Reddit all the time.
> StackOverflow reports that emacs fell from 5.2% to 4.1% from 2016 -2018, which seems large in percentage terms. Does it feel like usage is falling off?
I wouldn't trust stats from a single silo. I can talk about my interaction with SO (or https://emacs.stackexchange.com).
I was really active on SO/emacs.SE many years ago, especially when emacs.SE launched.
I'll be honest that that really helped me with my #emacsLisp skills. But then I switched to mainly using the gnu-emacs-help and emacs-devel mailing lists. So those stats are missing out votes from folks like me who don't use SO any more.
Also #Emacs has (one of the?) best introspective documentation (`C-h f`, `C-h v`, Info manuals, ..).
So the Emacs users who learn how to use that help would also not be active on SO to know that a poll is going on.
> And how is John Wiegley as the new maintainer?
I think that it's mainly Eli Zaretskii who is the actual maintainer of the duo. I haven't seen John Wiegley on emacs-devel for a long time, and also I don't see him making any commits. It had been Eli all through even before JW got elected as the maintainer.
> I think that it's mainly Eli Zaretskii who is the actual maintainer of the duo.
Yeah, I realized that after my original toot. I was basing my knowledge on the latest episode of Emacs Chat (http://sachachua.com/blog/2015/12/2015-12-10-emacs-chat-john-wiegley-maintaining-emacs-can-help/) but that is significantly out-of-date and I didn't realize that things had changed.
(Related: any tips for more recently updated Emacs podcasts?)
Though there are many YouTube channels; you can find them on show up occasionally on https://www.reddit.com/r/emacs .
From what I have seen, the video series by Mike Zamansky is the most active: https://cestlaz.github.io/stories/emacs/ .
@rakhim : We need more podcasts on your channel! :)
> I wouldn't trust stats from a single silo.
I'm not looking at the StackOverflow question/answer numbers (though I have in the past for other reasons) but in their developer survey. True, they advertise the survey on SO, so there is *some* bias towards site users. But they make at least an attempt to get respondents through other channels as well. They discuss their methodology here: https://insights.stackoverflow.com/survey/2018/#methodology
> I'm not looking at the StackOverflow question/answer numbers
I meant to say that there would be many people like me who don't visit any of SO channels any more, and so would have missed that poll.
If you follow the mailing lists, you will see the amount of activity there .. not just in terms of people asking for help but also in discussions related to emacs devel.
@kaushalmodi @codesections also the number of people on the Emacs IRC channel keeps going up... it’s been a while since I did the last one so data is from 2016: https://www.emacswiki.org/emacs/EmacsChannel#IRC_Client_Stats
Fosstodon is a Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.