Nedit is an orphan. How to revive it?

52midnight

Further to my previous post seeking a solution to a problem with Nedit:
https://forums.gentoo.org/viewtopic-t-1003478.html
I've posted to the 'discuss' list on the official website:
http://www.nedit.org/
sent an email to the email address provided:
http://www.nedit.org/contact/
and telephoned the Sydney, Australia number provided which responds with a "not connected" message. I therefore conclude that Nedit is now an orphan and has been abandoned by its maintainer (maintainers?).

Not sure what to do, or where this post belongs. I've been using Nedit for more than a decade and it's far better than any alternatives I've tried. I'd guess that its out-dated interface deters new users and has seen it relegated to 'legacy software'. IMO it deserves ongoing support, and could easily be forked to develop a superior and up-to-date programming environment - but please leave the functionality of the original 'as is' even if bindings to a new widget set are provided. The new generation of scripters are far too 'click-happy' for an old keyboard-only coder like me.

Any interest?

dataking · Apprentice Joined: 20 Apr 2005 Posts: 251

Download the source; upload to github; advertise as you see appropriate?

I don't use nedit, nor am I familiar with it, but I am a commandline warrior, so I sympathize with your plight. If I spoke C/C++ (assuming the source is that), and I used the product, I might offer to host it and guide development, but I don't. So I can only offer some cheerleading.

Sounds like it's time to learn vim, emacs, or whatever suits your fancy.
_________________
-= the D@7@k|n& =-

52midnight · Posted: Tue Nov 18, 2014 7:51 am Post subject:

dataking · Apprentice Joined: 20 Apr 2005 Posts: 251

dataking · Apprentice Joined: 20 Apr 2005 Posts: 251

Actually I just google'd nedit, and realized it's a graphical editor. /shrug My bad, taking "keyboard" to mean "command line".

According to Wikipedia: "According to the projects news page the source repository has been converted from cvs to git in September 2014.[3]". Maybe it's not entirely dead? Someone moved it to git, and I'm guessing it wasn't you?
_________________
-= the D@7@k|n& =-

52midnight · Posted: Tue Nov 18, 2014 7:48 pm Post subject:

NeddySeagoon · Posted: Tue Nov 18, 2014 9:39 pm Post subject:

52midnight,

nedit-5.5_p20110116-r2.ebuild is in the tree. That seems to be the latest release too.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

52midnight · Posted: Tue Nov 18, 2014 10:14 pm Post subject:

52midnight · Posted: Fri Nov 21, 2014 10:03 pm Post subject:

After a bit of poking around and sending emails to whomever, the following arrived:

Respondent #1 (private address):

steveL

52midnight · Posted: Sat Nov 22, 2014 7:18 pm Post subject:

Thanks for the interest.

> Yeah, though not in taking it to C++

Haven't checked the code although I think it's written in C.

> nor even a "modern" toolkit as that's just asking for hassle down the line imo.

I've used Gtk for a few quickies of my own. It's v well documented, mature, robust and pretty enough for my taste.

> Would be up for splitting it if it's feasible

Probably best to fork it, esp as the domain name is in limbo.

> so someone can use another toolkit if they want

Always wondered how feasible this is, keeping the executive code isolated from the GUI bindings. Would make an ineresting exercize of itself.

> If it's decent that shouldn't need much work, if any.

There are a few minor bugs, but it's pretty solid otherwise from my years of using it.

> might be something I'd find worth switching to

With any luck it should generate a new community of users.

> it's likely one I rejected for a crusty i/f back in the day.

Yep, many others would have done the same.

> Anyhow no real rush thankfully, so let's talk in the new year.

Agreed.

> By all means sketch out on paper the various modules you know about, and what bits you want changed.

OK. I'll get a copy of the source, sketch out some suggestions, and post back here in the new year.

steveL · Posted: Sun Nov 23, 2014 2:26 pm Post subject:

52midnight · Posted: Sun Nov 23, 2014 7:55 pm Post subject:

> Let's just prettify Motif instead.

I took a look through main() which is in nedit.c and half of the setup stuff is workarounds for bugs and "insanities" in Motif. Mind you, the date in the top comment block is ... now sit down, take a deep breath, and keep your medication handy ... 1991.

Not sure if this is still the case, but Motif was (is) proprietary so it might be pulling in lesstif. I suspect that prettifying Motif/lesstif would be a much bigger job than just cutting it out and substituting something more robust.

orionbelt · Apprentice Joined: 05 Apr 2006 Posts: 178

As a ~20-year user of nedit, i am very glad that someone has brought up the question of its survival. One problem that has not been mentioned in the previous discussion is how nedit does not deal with Unicode, and i believe also with TrueType and related fonts. Apparently this weakness is so tightly knit into nedit's code that it's not possible to address without essentially rewriting the editor... Although it might be that Unicode and cute fonts are not needed always and by everyone, i did need them (especially unicode), and that was the reason why i started using kate, although i still use nedit for special things.

The opinion that i formed years ago is that it would be great if nedit could somehow be maintained "in the freezer". I.e., if there can be no active development, at least make sure that it remains compilable under X.

52midnight · Posted: Mon Nov 24, 2014 1:29 am Post subject:

Many thanks for a most insightful post.

> how nedit does not deal with Unicode

There will be a number of similar issues. Should Windoze continue to be supported? (No interest personally) Is the server facility still relevant? Etc.

> it might be that Unicode and cute fonts are not needed always and by everyone

Case in point.

> at least make sure that it remains compilable under X.

Would be quite happy with this, esp if the few remaining bugs or whatever can be sorted.

> nedit seems to combine ...

Good promo. Read http://www.nedit.org/features/ for more.

> While Gtk may be lighter (is it?) than Qt, though also IMO uglier, i suspect ...

As a hardware guy I'm happy to defer to those with better knowledge of software issues.

> Having said all this, i should emphasize that i do not see nedit as the end-all be-all of editors!

Quite so, but I'm increasingly of the opinion that the mobile/gadget/InternetOfThings areas are drawing away so much vital manpower that us ordinary folk may soon be starved of it. What are the chances of a "free as in beer" rewrite of Nedit today, much less a new editor as good or better? Three, I'd say ...

> There are features that i do miss on nedit

I'd like to see custom keybindings, though this is far lower priority than just keeping it current. When you're on a good thing ...

52midnight · Posted: Mon Nov 24, 2014 2:07 am Post subject:

As an addendum, I'd like to combine two of orionbelt's statements into a pertinent suggestion:

> Decoupling the core code from the GUI toolkit would probably be the best option if it is doable

> i do not see nedit as the end-all be-all of editors!

Let's assume that we've all written enough editors to be tired of re-inventing the wheel, but agree that editors must continually be updated to accomodate new developments. Can we therefore devise a generalized schema that describes ALL editors - a sort of Backus-Naur Editor Formalism. For those unacquainted, see:

http://www.princeton.edu/~achaney/tmve/wiki100k/docs/Backus?Naur_Form.html

For starters I'd suggest that a key file could contain simple custom DEFINEs or MACROs to substitute bindings for different toolkits, e.g:

#define restoreInsaneVirtualKeyBindings(unsigned char* bindings) useISO-8859-1(unsigned char* bindings)

In this scenario, every conforming editor would consist of three parts: the executive code, the GUI bindings, and a substitution table linking them. I hasten to add that this is just fodder for private consideration, and should not in any way detract effort from just KEEPING NEDIT ALIVE.