Hacker News new | ask | show | jobs
by MarcScott 4435 days ago
There seems to be a lot of text editors being showcased on HN at the moment. While the engineering involved in each of them is remarkable, and way beyond what I could imagine accomplishing, I've yet to be convinced that any problems are being solved. Nothing I've seen has yet convinced me to move away from my emacs/git based workflow.

Maybe they are simply CV showcases, in which case this is an impressive piece of software, and congratulations to the developer, but you've yet to convince me to move away from 1970s technology.

6 comments

Most of them are just CodeMirror / Ace stuffed that takes 90% of the pain of writing anything complicated in that domain, really. The CoreMirror author should take a fee for each product based on his lib, would become a millionaire( same guy who wrote TernJs , Eloquent Javascript , and few others, this guy's insane js developper, not sure how he manages to write this huge libraries in plane javascript and vim ).
I've heard of Codemirror but went and took a look again after reading this comment - I was surprised to see that Brackets, Lighttable and the Chrome DevTools source editor are built on top of it.
Completely, this is why I love how modular that Javascript ecosystem is. Building a complex piece of software only takes 450 lines of glue [1]. Node-webkit makes building desktop applications a snap. The only bummer is it's slow, when I type I'll notice that 3 characters appear at once, it's distracting.

[1] https://github.com/jawerty/Hyro/blob/master/js/hyro.js

Many of us go our own way for maximum flexibility. You'd be surprised how easy it is to write a modern live editor as long you are given the right tools for font rendering and extracting font metrics. The hard part is achieving the liveness, though.
I understand your feelings but this isn't a heavily planned startup i've developed, simply a tool I made since I thought it would be cool to make (also an experiment with node-webkit).
I often have exactly the same reaction - the main question being: why? And let me highlight that I am also impressed with the engineering skills required by this (and similar) projects.

The speculative answer that comes to my mind is that engineering is the 'easier' part of coming with a new product. The concept is the 'difficult' part. I put these qualifiers in quotes because there is nothing easy about the engineering, but it comes easier to people that are already working with technology. They know how to make things. Deciding what to make is not an engineering question, but a business (or society) one.

I suspect that this is the reason for quite a few failed startups. They build things that they can, but not necessarily should.

But, again, this is not a criticism of this project, just adjacent discussion.

I have a different problem. All of these new fangled cloud-based html text-editors / IDEs are kinda .. dinky. They're toys. It seems like everyone is tripping over themselves to include as few features as possible and hope they can sell them for $3.99. Where are those editors that are worth a $500 price tag?
"Where are those editors that are worth a $500 price tag?" - they would need much more code and thus require a maintainable programming language.
> ...and thus require a maintainable programming language.

Hum, bit of a lazy dig at JS there.

Also tell that to the Brackets team [1]. It has a pretty decent feature-set.

[1] http://brackets.io/

I have to admit, that is some quality JS code.
My similar product (http://liveditor.com) mainly solves this problem: After tweaking css using Firebug, having to manually copy the changes back to the source code is time-consuming and error-prone.

And LIVEditor's solution is integrating a Firebug-inspired html inspector into a full-featured code editor, of course plus a Chromium-powered real-time preview!

It's nice and simple, I like it. I can't see leaving Notepad++ on Windows or Geany on GNU/Linux for it, but I do like the concept. I think it would be good for someone just learning to use HTML, as they can see immediate results.