I certainly didn't expect the team would be "revisiting some of the foundational ideas of computing" so soon.
LightTable right now is a proof of concept. A very good one, but still — it has a long way to go until it can become an everyday tool. Let's not leave it hanging there. If you continue on this path, you will leave behind a trail of half-baked tools, each one a great concept, but none of them actually useful.
FWIW, I agree with you. This is our path, though, and the funding and the people we've brought on to work with us prove that. We're locked in to Eve and we think it will ultimately have far greater and for more lasting impact than LT ever could.
So the best I can say is that we're all in on Eve (both personally and now as a company) and I think that's very important for this to have a chance to succeed.
I backed Light Table and enjoy it (though it is nowhere near my daily editor) but I'm excited about the shift in focus. Sometimes great ideas do not pan out to great products for a variety of reasons, and it would be fiscal suicide for them to keep the focus on Light Table if the team truly does not believe they will be able to reach the product/market fit they were hoping for. This is business, not charity after all.
To put it another way, Eve sounds like a really interesting solution to a hard problem and I'm really excited that this team is trying to tackle it. Stoked to see what the future brings.
I can understand having ran out of the original kickstarter money, but after enteing Ycombinator I'd expect them to still have resources to avoid being forced to pivot
(trivia/disclaimer: I'm a kickstarter backer and Light Table was my main editor for some months)
And, while Light Table is a polished little editor, it's hard to call Light Table an IDE without offering an apology. Light Table certainly is not what was promised.
In fact, Light Table was my first experience really being burned by a Kickstarter project. Nice that Chris Granger wants to reimagine computing, but he should finish what he starts.
Who would gift his crew with $2.3M given their track record? Atom, which did not promise an IDE has done more work on it's editor in a shorter time than Chris and crew. Buyer beware!
It's a company full of smart people backed by smart investors, if there was a way to make LT work at scale I'm sure they would still be working on it.
I was taught that when someone gives you even one dollar in consideration of something that you've promised, that taking the money is tantamount to an oath.
Light Table essentially proves the lie, since it is pretty far from the video. From where I sit, Chris and his team broke their oath and I won't be burned twice.
If Andreesen Horowitz believe that computing can be revolutionized for $2.3M, it's their cash to burn. Not surprised at all that they didn't turn to Kickstarter for funding.
The word is out.
I was very much interested in LightTable. If I had given money to the kickstarter fund I would be very ticked to find out that those funds were being redirected to a different project (and I would not give funds to those individuals again).
I am also trying to be helpful by pointing this out (not just taking a cheap shot).
Also, my gut reaction to this is "LightTable is effectively dead". Again, not trying to be mean -- just basing this on a long history of following the industry.
All that being said, it does seem that the LightTable folks are trying to be transparent about all of this, and that much is refreshing.
If they are successful in producing something very useful, my hat is off to them. But I have to admit that it sounds kind of "pie in the sky" to me in this announcement. Please feel free to prove me wrong :-D
It would be a shame for that to be the case. The amount given was indeed much more than reasonably used up if it supported 3 people for 2.5 years (especially in SF).
While that's good to know, I think that the implicit upside gain hoped for from Kickstarter that underlies a lot of decisions to fund through it (especially for software) is that acheiving funding means not just that the thing gets delivered, but that an effort is made to build a sustainable business model is built around it and support the user community. The perception that, having delivered the kickstarted product and spent a couple years on it the team would abandon it and move on to a very different kind of focus (even if in the same broadly drawn domain) understandably reduces the perception that that kind of upside potential is around for a future effort.
Now, one can argue that that kind of upside expectation is somewhat irrational in the first place, of course.
If it truly is "a great IDE" and "95% complete in every corner", yet it dies because the users (who all happen to be coders!) don't rally around it and contribute that last 5%, what does that mean?
And just because all of the users are devs doesn't mean they're devs who have time to dedicate to building an IDE while they could use something else like KDevelop, PyCharm, Sublime, vim, or hell even VS.
Any single tool is going to be an incremental improvement (at best) on programming, because the source of the issues is really the model itself. My suspicion is that programming is going to change substantially over the next decade, because while we've been worried about machine learning, rails, and hadoop, the notion of the machine changed out from under us. We're already starting to hit a wall as a result of that - reasoning about highly concurrent systems with the programming models we have now is nearly impossible. If we continue pushing LT for that way of thinking, it's unlikely to make much of a difference. Instead, there's an opportunity to embrace that change and take the ideas of LT to their logical conclusion: what would a stack built for today's machine that is meant to be used by humans look like? We stumbled on a solution as we did more and more user testing of LT and in the interest of self preservation we ran with it.
More practically, though, incremental improvements are hard to sell - not just to VCistan, but to programmers as well. Hell, even orders of magnitude improvements are hard to sell to programmers. We'd have no way of sustaining ourselves if we couldn't find some avenue to eventually turn this into a business - we looked and couldn't really find a reasonably quick way to do that. Matching the millions of man hours that have been poured into vim/emacs and so on is just not in the cards for a couple of guys with not much funding. Plus it seems programmers are increasingly less interested in paying for anything in general, so that didn't make our prospects all that attractive. With Eve there's a much clearer path towards creating a sustainable business that is able to continue towards the ridiculous goal we set out with from the beginning: to empower the world's creators.
We use LT every day. Lots of other people do too. It fills an important part of the ecosystem and we hope it continues to infect other tools and organizations who have significantly greater resources than a couple of guys in SF. But either way we will do our best to make sure that it continues. Unfortunately, though, it just can't be our primary focus - both because I think that would be focusing on the wrong part of the problem and because at the end of the day we have to pay ourselves and there's just no way for us to do that.
I still think that LT is going to slowly peter out, as there isn't enough momentum to dethrone existing IDEs / upcoming actively developed contenders.
That said, I think this level of transparency about where your collective heads are at is great. While some people may still be annoyed (particularly those who donated to the kickstarter) I applaud your willingness to change paths and do what you think needs to be done to succeed.
I'm excited to start playing with Eve.
A lot of people are interested in the project, would like to contribute to it, and would be happy to take over where you left off. However, this is difficult to do due to lack of documentation and any support from yourself.
You can read the experiences of one person trying to contribute here (https://groups.google.com/forum/#!topic/light-table-discussi...) and here (https://github.com/JonyEpsilon/gorilla-repl/issues/132).
Since you're not going to be actively working on the project, why not involve the community find individuals or an organization that would continue developing and maintaining it.
Surely, this is not too much to ask and it would be a responsible way to transition from the project.
I will certainly not stand in the way of the community and we're more than happy to help maintain.
Simply dropping the code on GitHub and saying that you're glad for people to get involved is not a sufficient level of support at this point. This might've worked when the project was younger and smaller, but it's just too big for people to be able to dive into now. The links I provided are a prime example of the problem.
Since you found the time to make a blog post about Eve, surely you could also make a post stating that you're looking to hand over Light Table and you'd like to help facilitate the transition.
a) More "traditional" languages aren't going anywhere, at least for the next few decades, however successful Eve is.
b) The idea of really connecting to the running program has a huge amount of potential, even in these languages. Less potential than Eve? Perhaps, but I don't believe for a second that we've hit that wall yet (and I wholly intend on proving that with continued work on the Julia plugin).
Leaving Light Table unfinished would be a travesty, even if it is only an incremental improvement – but it sounds like you'll at least treat it as an important side project, which is something. Best of luck with Eve, anyway.
I fully support your ideas with Eve.
Every time I dig into how Xerox PARC systems worked, my own experience with Smalltalk and Oberon, and then look for what we have, I always think we could have had so much better developer experiences if those technologies had succeeded in the market.
Good luck with your efforts.
Whenever I've used it I've felt like a small amount of effort in those aspects will allow it to continue to progress in the hands of the plugin community.
Looks like it's time to dive into Emacs...
Like Unix you can't grok all of it (ever), but you won't ever need all of it. Just learn more of it as you go along. See it as a trip :-)
And if you need your Emacs Clojure development environment right now, have someone make a VM for you.
...and customized it enough to have an environment comparable to what other editors ship out of the box
wc -l ~/.emacs.d/init.el
241 /home/dario/.emacs.d/init.el
241 lines of emacs lisp, and I'm still unsatisfied with the current state of my emacs ls -l ~/.emacs.d/emacs.org
-rw-r--r-- 1 user user 67417 Oct 1 17:01 ~/.emacs.d/emacs.org
And after 15 years I'm getting finally somewhat satisfied. However, the alternatives were always worse.The ctrl+spacebar and search way that Light Table works, while not the fastest, was still preferable to holding down multiple modifier keys.
All that being said, the source of my hand pain might actually have to do with my neck and back, as working on those areas has substantially relieved my pain. So perhaps that will no longer be a barrier to emacs usage in a few months. :)
I've never used the defaults[1] for long. First by making my own custom keybindings, then by making keybindings compatible to what most others applications use and a couple of years ago by switching to evil-mode. I already had vi/Vim experience so that was pretty optimal.
[1] Except some really basic ones that I also use in the shell: C-a, C-e, C-r, etc.
and there's also vim-fireplace:
https://github.com/tpope/vim-fireplace
By all means, if you don't want to use emacs then you don't have to!
This is a simple truth.
For my part I think they've delivered on the spirit of their Kickstarter. They released a very usable editor that does some amazing shit. It doesn't have every feature in the Kickstarter video, but that's because reality kicked in and they had to prioritize, and found that some of the features didn't work out as well when put to other users.
LT fills multiple spaces in the area of IDEs:
- An extremely-programmable IDE (there's no other one like it)
- A lightweight generic IDE (there's REALLY no other one like it). I almost wouldn't call it an IDE but it's more than a text editor... it's right in the middle of the two and it's great to have something fill that gap, especially for languages that don't always need full-blown IDE support (like python and clojure)
It'd take a long time to explain the differences but emacs is far from lightweight, it's a beast. LT is very barebones. And LT is written in HTML/CSS/CLJS which makes it stylable and programmable as such, a real advantage over emacs.
Emacs is an IDE, first and foremost, to Emacs Lisp itself. Other Lisps are close. It's an excellent editors for other languages, and a good enough IDE for some languages. See my C/C++ guide in Emacs: http://tuhdo.github.io/c-ide.html.
Fyi, people wrote parsers, chat and email clients in Emacs that LT is still dreaming.
etags -- it's just an nightmare, full stop. It only gets you where you want to go half the time at most. And then because the GUI is so primitive you can't do basic things like click on a symbol and have it jump you to the definition, you're stuck using the keyboard interface designed for work in a terminal.
An etags that actually worked properly is one of the most basic things that could improve my productivity monumentally.
CEDET -- last used about 5 years ago, gave up because on larger files/projects it would just hang my editor for a few seconds as it processed.
Variable/symbol completion -- see etags.
I'm not doing a very good job communicating my dislikes / pain points because I'm not very good at communicating. Please do not respond with a "but these aren't problems because..." because they are problems for me, I just do a poor job of communicating.
I gave up on emacs for Java development for the reasons you cited, but I still use it for REPL style development (Jython, Ruby Motion, XQuery, as a few current examples).
And use GNU Global instead of etags. Finally, keyboard boost productivity many times over the mouse.
Google Chrome, or any modern browser for that matter, has a much better way of doing things. Imagine for a moment that I was browsing and editing code in Google Chrome instead. A symbol would be a link, and when I ctrl+clicked on it, it would open in a new frame. That would be an interface I could get behind.
For completion, it ought to work like the IDE I used back when I was writing Scala. A little tooltip would pop up with completions. You could scroll through them with the mouse or keyboard, and click or ENTER the one you want and it would paste in. Or you could ignore it completely and continue typing. This sort of UI is so much better than popping up a new buffer in the frame. But in Emacs UI, everything is a buffer and there are only three places they can (or usually do) go: in a new frame, in a new window, or in the minibuffer. Only that last one is usable.
GNU global is not a solution. It produces marginally better output than etags, but still requires you to actually run it whenever you make changes to the codebase. Or at least it did back when I tried it as an etags replacement.
There's nothing elite about the keyboard. It's just the keyboard enable higher productivity when it comes to write code.
For completion, it ought to work like the IDE I used back when I was writing Scala. A little tooltip would pop up with completions. You could scroll through them with the mouse or keyboard, and click or ENTER the one you want and it would paste in. Or you could ignore it completely and continue typing. This sort of UI is so much better than popping up a new buffer in the frame. But in Emacs UI, everything is a buffer and there are only three places they can (or usually do) go: in a new frame, in a new window, or in the minibuffer. Only that last one is usable.
In Emacs you have tooltip as well. For example, company-mode shows a completion popup like this: http://tuhdo.github.io/static/c-ide/company-clang-company-gt...
Completion using a window buffer is still vastly superior because it can handle large collections. For example, here is how I filter completion candidates with Helm: http://tuhdo.github.io/static/c-ide/semantic-boost-demo.gif. Helm works even if you have 99999 candidates (not joking) to filter out. Here is how I quickly filter to a file and enter the file in Linux kernel source tree: http://tuhdo.github.io/static/helm_projectile.gif , very quick and simple; in contrast, the file search dialog box in typical IDEs is very complex with many checkboxes and text fields. How do you cope with completion popup with contains thousands of candidates in it?
GNU global is not a solution. It produces marginally better output than etags, but still requires you to actually run it whenever you make changes to the codebase. Or at least it did back when I tried it as an etags replacement.
No, ggtags: https://github.com/leoliu/ggtags or helm-gtags: https://github.com/syohex/emacs-helm-gtags automatically updates your tag database when you save your file. The tags produced by GNU Global is A LOT better than ctags for C/C++.
Maybe you can look at a few demo screenshots in my C/C++ guide: http://tuhdo.github.io/c-ide.html
I mean Light Table is a great IDE but it just needs so much done. It feels like 95% complete in every corner. The vim keybindings are buggy, Python plugin lacks proper virtualenv/custom python support, performance is not quite there either... I forget all the issues I encountered back when I used it but there has been very few releases since and it's super concerning to just see what was once an extremely active project slowly go in maintenance mode.
Please, quell my fears. You guys put together an amazing Kickstarter and used to put out so much in terms of updates and feature updates; I don't see that anymore.
With all that said, best of luck with Eve, sounds awesome.
... Edit: I think these graphs are pretty telling. https://github.com/LightTable/LightTable/graphs/code-frequen...