Hacker News new | past | comments | ask | show | jobs | submit login
Light Table Is On Kickstarter (kickstarter.com)
619 points by ibdknox on Apr 17, 2012 | hide | past | web | favorite | 215 comments

Just added this update:

EDIT: Also, these changes will take effect for everyone who has already pledged at these levels. Though if you could, please switch - if the old ones have 0 pledges I can delete them.

Unfortunately I can't change the pricing structure in place, but based on the feedback we've gotten so far I've added a set of new rewards - all of which are better than their previous ones. :)

$15 - A license and acknowledgement on the contributor list

$50 - A license and early beta access

$100 - A license, early beta access, and participation in feedback rounds

$200 - Two licenses, pre-beta access, and participation in feedback rounds

All the higher rewards will also of course include participation in the feedback rounds. Ultimately, this should address people's concerns of buying something before it exists, but providing the opportunity to be a part of it at a lower price point.

Awesome, you sir have just gotten a new pledge! I am much more confident knowing that you'll listen to feedback.

Just a thought that popped into my head. Assuming that the product issued in the end is going to be the same for everybody. And assume that the highest contributor that gets to put in the feedback generally improves the software. This creates a free loading problem for those who contributed less (myself included) for the same product. Even if it's a positive externality for the general backers, I wonder if there is a way for Kickstarter to properly compensate for such externality.

Thank you for adjusting the price!.. You're forcing me to put my money where my mouth is :)

I've also sent KS mail about seeing if they can merge the rewards. We'll see what happens.

Thank you, I don't really have a use for Light Table (but it's a nice concept), but after I read this update, I instantly signed up, if not only just because you made noir ;)

Thanks for the new reward levels. I don't know how to code but $15/license for the opportunity to learn with lighttable is a steal.

Much better now. I pledged 50$ for the beta access and license. Good luck!

Does "early beta access" mean we get also get early access to the open-parts of the source? I'm fascinated by light table and have some ideas to take it to some underserved markets.

The timelines are all messed up now. What's the difference between 'early' and 'pre' beta? Are these still later than the $500 'absolute earliest user testing' pledge?

If licences are now down to $15 - shouldn't all the pledges that contained multiple licences be adjusted as well? e.g., the $500 pledge included $200 worth of licences before, but now they are only valued at $60.

I understand KickStarter doesn't allow you to change existing rewards, but this is turning into a bit of a confusing mess...

I agree - I'm pretty disappointed in kickstarter's tools.

Here's a rough guideline:

Earliest access - August 2012

Pre beta - October 2012

early beta - Decemeber 2012

Official Beta - late January 2013

Launch - May 2013

Fantastic. I just pledged.

I think the kickstarter page could be improved.

More details about the money and why you need it, what our donation will accomplish aside from the vague message about how more people get things done faster. How much are you paying yourself? How many other people?

More information about the end product (people are asking about OS for example), the timeline, and you and your qualifications to run the project. I gather you have been programming for a long time and work at Microsoft, but that doesn't prove you'll be able to deliver. For all I know you're a famous hacker, but not knowing that, not knowing if I can use the end product or when I will get it, and not being given a simple, compelling argument (rather than a detailed description of the idea) makes me less interested in participating.

That said I love the idea and really hope you are successful!

I just added this:

And why should we believe you can deliver?

I've been programming professionally for 8 years, during which I've built websites large and small (puma.com, newbalance.com, etc), helped design the future of Visual Studio, and released numerous open source libraries and frameworks. While at Microsoft I was the Program Manager for the C# and VB IDE where I spent countless hours behind a one way mirror learning how people develop things. Since then I've steeped myself in the world of startups and OSS. I worked with the guys at ReadyForZero to build readyforzero.com, created the Noir web framework, built the SQL abstraction layer Korma, and released a host of ClojureScript libraries to make client side development a breeze - many of which are now featured in the canonical books for Clojure. Even more recently, I built Bret Victor's live game editor after watching his inspiring "Inventing on Principle" talk. I have a history of building useful software and a hunger for making developers more efficient, as well as experience leading and shaping the the future of a billion dollar project.

FYI I personally found your experience on the MS Visual Studio team the most convincing credential, may I humbly suggest that you lead with that whenever possible.

PS. Good luck.

I agree, this makes me the most confident in the final product.

Definitely an improvement! Great job.

To your question: Chris is very talented and well known in the Clojure space: I use a few of his open source projects, good stuff.

Yes, but that should be on the Kickstarter page.

I really want to back this but $100 is pretty steep for beta access. Even $50 is a little pricey for a tool that I like the goals of, but find it unlikely that I'll use it for a commercial product anytime soon. Any chance you could move around the reward tiers?

Maybe I'm just spoiled from spending $15-$20 on Kickstarter games and getting beta access, but as long as Light Table looks like an awesome toy instead of something I can build commercial software with it's hard to justify dropping $100 on it. That said I'll probably throw in $50.

For developer tools you use every day, I think $100 is a relatively small amount and in line with what a lot of text editors and IDEs charge for (and less than many of them).

For a real product, yes. For a product that doesn't exist yet, may turn to vaporware, may be buggy, may end up poorly designed, may be unusably slow, might not be able to operate as well as described, etc etc, $100 might seem a bit steep.

Perhaps, but part of this is demonstrating the demand for a product of this kind. If Chris reaches the goal for funding, but fails to release a product, hopefully the fact he managed to raise that much cash will indicate to someone else - Microsoft, IntelliJ, Eclipse developers - that there is real money to be made in reinventing the IDE.

I don't see it as chipping in $100 for Lighttable itself (which I do want), I see it as showing there is a very real demand for new thinking in IDE-land.

People spend $2 a day for coffee. Two movie tickets in NYC can cost $30, or more. I think $50 is pretty cheap. Even $100 is not that bad. If it saves you an hour a month, it has already paid for itself. The coffee, well, not really.

You say that, but would you forsake coffee for 50 days for an IDE?

It's only 25 days if you can wait until after the beta. :-)

It's only $100 so there's no need to forsake anything. I bought a personal license for IntelliJ IDEA, for example, and I upgrade to every new version. I believe in buying software that makes me more productive.

I find it kind of strange that people in the software business have such a hard time paying for software. I hope this doesn't rub off on the general population.

That's not how you should think about it. A programmer's salary is say 50k. If this IDE makes you 5% more productive, then it is worth paying $200 per MONTH. If anything, this is vastly under-priced.

But this is a beta product. No demo and nothing to base your decision on but a video. $100 is a lot to bet for a v1 tool that you might never wind up using.

And that's the interesting thing about Kickstarter – am I backing it just for the raw value of the product or because I want to see it actually get made? If the latter, am I willing to pay more?

In this case, personally, I am.

Based on how the software world works, I'm not willing to pay that much money to take a risk on a software project. Failure is way too commonplace. There's lots of software projects many people wanted to see made that were a disaster once actually completed. See for example Duke Nukem Forever

And that's a completely fair point of view. In my case, I want this project to be real badly enough that I'm willing to put some money behind it... but I recognize it for what it is. A work in progress and a hope for a good tool.

I agree. The issue is I have no idea if I'd use this everyday, or just once or twice to try it out. I don't see any company I'd work for adopting, but if I knew it would be anywhere near as useful as some of the free tools I use for side projects (Processing, Flashdevelop, etc.), I wouldn't hesitate to back it for $100. I guess that's the risk Kickstarter and being an early adopter in general.

Agree as well, too risky for $100 that there isn't even a alpha to try out. $50 and I may bite.

So the product will be open source, but you need a license to download it? So its open source, but not "free". Are there any well known licenses that support this?

I take it that it also means that the license will not allow for you to copy / redistribute Light Table? Does it still really qualify as open source then?

Will they allow for contributors to the core of the product and if so, what rights do contributors need to sign away?

That said, I think its an awesome project!

In some ways it sounds like the Microsoft "Shared Source" where you are able to view the code for some of their commercial products for debugging purposes but not redistribute it.

One concern is they mention that it will make use of existing open source software; presumably they mean more BSD-style code or LGPL code rather than GPL code since the license they are proposing seems incompatible with the GPL.

This is pretty important to me too. $50 isn't much at all for something as important as an editor but I don't think I'd consider switching if it wasn't GPL or MIT (or similar).

But of course it's important that they be able to make money too, I wonder how that could work.

Will pledge my $50 later today, I'll probably never switch from Emacs but I do want to see where this goes! Id make it 100 though if they promised to (properly) open source the core and the language plugins. Maybe that could be one of the things promised if they exceed their funding target?

Atlassian uses a model similar to what you're describing. If you buy a license, you get access to the source code, but you can't redistribute that source code. I don't know if that can really be considered (as the author refers it) "open source" though...

The strength of his past contributions alone is enough to convince me to donate. You can't throw a football in Clojure land without hitting a high-quality contribution of his.

Does anyone else find it awkward to back a project where the person running the project has never backed another project? I'm stoked about this project, It's just a feeling I've noticed consistently when looking at Kickstarter campaigns.

kickstarters are not the sole way people contribute to projects and communities. ibdknox has created some very popular (in the clojure community anyway) software (noir, korma etc) that a lot of people build on top of. Surely this should count for something.

Oh for sure, I was just curious about people and the platform. It's the only listed metric below a creators name, so if you didn't know who he was, it would be hard to verify. That being said, I've already donated to this.

You shouldn't be downvoted. It's a completely legitimate point to make. Kickstarters from people who don't contribute back is something that happens all the time.

In this particular case this one is from someone who contributes to the community in other ways, but that's certainly not always the case.

Just added this to the description:

If we hit $300k, Python will be the third language to be supported out of the gate.

Get it out first, if you hit $300K, put Python next on the list, but don't delay release until it's in there. It's just good sense to get something out sooner regardless if you have the finances to support more.

What would you have to raise to promise Lua support ;)? I imagine Ruby would come in at $400K-$500K.

Curious, what do you use lua for? Presumably embedded in some fashion? I'd imagine the use case will change drastically based on the target.

+1 for Lua

I'd donate $500 right now if Lua support will be available in v1.

Edit: $500 (originally $100)

I rather doubt there will be that much point in that. Based on what he has spoken of the architecture, doing 3rd-party language support should be extraordinarily easy. Putting more languages into v1 will really only push it back -- the actual lua support should probably be done by someone with experience hacking lua.


I'm not upset that higher pledge tiers don't include the rewards of previous tiers. This is not always cost effective, especially when you factor in T-shirt production and whatnot. I do believe however that the higher levels should give more value to the backer, where value can be reasonable calculated. Specifically, I take issue with the higher of these two tiers:

> PLEDGE $1,000 OR MORE - 10 licenses of Light Table and acknowledgement as a backer of the project.

> PLEDGE $5,000 OR MORE - 25 licenses of Light Table and acknowledgement as a sponsor of the project.

How can a pledging 5x the amount of the previous tier give you only 2.5x the value?

Well, you get your logo on the home page for a year, which could be very good advertising if the project becomes big. However, I have no clue how much the advertising would actually be valued for a typical company.

You get to be a 'sponsor'. That's obviously twice as good as a 'backer'.

Honestly, I have no idea. In the end, these levels are primarily oriented at companies or other groups (education?), not individuals. What's one person going to do with 25 licenses?

I’d like to know what he plans to do with $200,000. Is he staffing up a company?

It'll take more than just me. It'll be a team effort.

$200k feels like a pretty random number -- and maybe even a bit high. Draw people into your story and elaborate on where the $200k will go. What kind of a team are you going to put together? A team of 8? A team of 2? Water-front office space? Time share private helicopter?

This is the moment to really create momentum for the project -- if I were in your shoes, I'd spend a bunch of time today really refining the project details on Kickstarter. (Other successful Kickstarter projects might provide some great ideas.)

Cool, I have no doubt it’ll require a team. Offer a few more specifics on the Kickstarter page. How many people? Time frame?

I say this to help your efforts – more specifics mean more likely support, imho.

The goal for this one was 400k :


The point being, that you should lower the goal and the main milestone. Esp. as long as you don't have a team to show... if there's interest it can go up many times over your goal anyway. People would ignore the goals if they can get something compelling from their individual deals.

$200k would be a good start, IMHO. I have developed simpler apps that have cost far more than that. I am guessing this is probably going to be at least 2 years of effort if its just one man, and if a bigger team then he will definitely need the funds, especially to hire quality devs.

I'd like to know too, but I'm not skeptical of the 200k figure. If this is going to work then it's not going to be a solo effort, so let's say one or two more devs and someone to do QA or support or design or whatever and this budget is looking pretty tight.

Without disagreeing with the prevailing opinions on the reward rankings, may I observe that it's interesting how Kickstarter has transformed donations for innovation into a shopping cart experience?

So many folks in this thread, myself included, have adopted a stance of "I don't know if pledging this much is worth it!"

Well, we don't have to, right? We can just pledge $1 if we like it and want to help a bit.

Except we're talking about this product as though it is on sale and priced too high.

[edit]: Cleared up language.

it's interesting how Kickstarter has transformed donations for innovation into a shopping cart experience

I've been going back and forth about this, and I think the phrasing of the particular Kickstarter project controls whether it's more a donation or shopping cart experience. Think about the conversation like this:

- "Hey, can you chip in some money so I can spend a bunch of time making a really cool piece of software?" - "Sure, I'd love to! I can spare $25. When do I get to play with your build?" - "Oh, $25 isn't enough for you to access the software. That's just a donation. But you can buy a copy when it comes out." - "..."

If the pitch is "chip in what you can, and we'll make a cool thing for everyone," then you set the stage for donations for innovation. If the pitch is "pay enough, and we'll give you a license to our cool thing," then you're offering a shopping cart experience. Which is fine, I use shopping carts all the time, but it's a different and less exciting thing.

This is maybe different with physical products -- if there's a cost per product, there obviously needs to be a minimum donation to get one. But with a virtual product, the reason you set a high donation threshold to get a copy is that you're selling a product rather than asking for a donation. Again, that's totally reasonable, but it's not as exciting to be part of.

As I keep thinking about it, I think there might be an uncanny valley effect here. If a software company I really like, like the Omni Group or Macromates, said they were doing presales of a revolutionary new product and needed me to buy in so it could get made, I would be psyched to do it. But this is halfway in between -- it's neither "donate to my cool project" nor "pre-order my cool product," but "donate to my cool product." I think that's where our gears start grinding a little.

I'm still up in the air about whether to donate to/purchase this one or not. Interesting stuff.

Clarification on the closed source plugins:

That bit of text was misleading, the plugins that would be closed source are not the ones that would result immediately from this project, but instead domain specific plugins that wouldn't have broad appeal. All the language plugins will be open.

I really think you should clarify the expected terms of the license.

I would gladly throw $50 to you if you released this under a MIT (or similar) license. Even if I never switch from vim, I feel my $50 would be well-spent bringing a (hopefully) awesome free IDE into the world.

If this isn't actually free software but only open source, I still might purchase a license for myself, but I think it's fair that this is mentioned upfront.

I think the $50 mark to get a license is a little steep.

Besides that, the video quality on the Kickstart home page is quite bad; I tried to see what was going on so I put it in full-screen mode and couldn't read anything properly.

I'll happily chip in some money, but having read recently about how Star Command's money burnt down (http://www.1up.com/news/star-command-kickstarter-funding-qui...) I suspect you might have shot yourself in the foot a bit by offering the t-shirts at such a cheap rate.

Either way, I've put some cash in the pot :).

I'm confused about what Kickstarter will accept for software projects and which ones they won't. My latest project was rejected, but it was also a developer app. What is the guideline for what they will accept?

I was hoping it'd be open sourced. On the other hand, maybe Chris will do amazing things if he has the ability to work full time on light table.


    Will it be open source?

    I'm a firm believer in open source software and open
    source technologies. I can guarantee you that Light Table 
    will be built on top of the technologies that are freely 
    available to us today. As such, I believe it only fair 
    that the core of Light Table be open sourced once it is 
    launched, while some of the plugins may remained closed 
    source. At some level, this is an experiment in how open 
    source and business can mix - it will be educational for 
    us all.

Yes, but the rewards also mention a license, that will apparently be based on 'pay what you think it's worth'. It all seems very vague to me and I'm not sure how it all will work. Might just be me though.

This is what investing is like.

Doesn't that just translate into "I believe in Open Source when it's stuff that I can use for free and as long as it doesn't take away from my revenue"? What kind of belief is that?

Open Source Business models have been under debate and development for a while now, making it seem like he needs to "experiment" here shows that he doesn't have much insight into the Ecosystem to begin with.

To me "once it's launched" means it's a cathedral design where no one will see what's going on with the project until some future date.

I'd prefer it if third-party contributions were allowed right away, and I'd be completely fine with only the core developers being paid. I think getting early feedback from users who care enough to contribute would be invaluable here.

even though he is not a politician, reading first 5 words that he believes in something bla bla bla, BUT. There is always that BUT.

Just answer yes or no, thats how simple it should be.

I completely agree, I noted it too.

Basically, what he is saying is "It's not going to be open source or free software." but in a way that makes it sound less bad.

It is - look at the questions at the bottom.

I hate when pledges don't include the lower rewards. How can I get a t-shirt and a license?

A few recent vets have indicated the issues with having rollups of previous tiers

See http://zefrank.tumblr.com/post/20122841731/kickstarter-post-...

"Be careful with waterfalling rewards (having each category include all previous ones) — We had t-shirts in too many tiers; and they ended up being one of the most expensive rewards"

This was the thing I was worried about. Both in terms of the cost in time, shipping, and materials. That being said, if people really want t-shirts we'll find another, better way to make that happen :D

Doing tshirts seems like a simple way to raise cash or reward people, until you've done tshirts for the first time. Making sure you have a quality product, handling inventory and fulfillment, if you've never managed physical products before, it can be overwhelming, and I've seen that bite people in the ass.

I pre-sold tshirts to fund a music news website about ten years ago. I only sold about 35 shirts, but given that I was in school and working two part time jobs in addition to keeping my site going as a hobby, I found even fulfilling just those 35 shirts to be way more work than I really wanted, and I haven't bothered selling shirts since.

Although after everything was mailed off, I did make enough to cover hosting for pretty decent amount of time.

Could the overwhelming aspects be automated? I can see where you would need to design the initial shirt and verify its quality, but isn't the rest just exporting a portion of your database and giving it to a t-shirt printer that ships? (or maybe they don't do that for you?)

Or is the problem that people are manually doing something that cafepress or zazzle provide might provide?

Of course, selling t-shirts for $30 should make a lot of that pain acceptable.

After subtracting the cost of the shirt, shipping, and your time, that leaves you with what, $10 each?

It's not terrible, but it's nowhere near as efficient as selling digital goods.

Should put something like "if we pass the X goal, then it will be truly free, no licensing holding progress back".

Then I would be excited to donate :)

I don't mind paying for the software. If it ends up being as cool as it looks, I'll happily part with the cash.

But I would love the promise of open source and not open core. Especially since the current description is so vague.

"some of the plugins may remained [sic] closed source"

What kinds of plugins? Every supported language is a plugin -- will Python or Ruby end up closed? Or the inline HTML rendering?

If Chris promised a true commitment to keeping this open, I'd promise to pay up. With Kickstarter, I could even prove it.

How about a KS reward that would give me the access to all everything Light Table related that the team builds?

Also note that even GPL allows you to sell the software and sell the code separately. Ditto for other licenses. I'd pay for the code. Not sure how viable that actually is, but it's something to consider.

Lastly, how's the development going to look like? Are the open source parts going to be done in the open (as most successful open source projects do) or will it be Android-like code dump on the release date?

I might have overlooked it, but it doesn't seem to say which OS Light Table will run on?

I would back it up for Linux, for example, but a Windows version wouldn't be useful for me.

I think it will be browser-based. He mentioned using codemirror in his post here: http://www.chris-granger.com/2012/04/15/light-tables-numbers...

It'll support windows/linux/mac

Thanks for the quick reply ibdknox! It might be useful to include it on the kickstarter page too?

This is too much. I've never seen an editor that runs well on the 3 of them. If you had said Mac or just Linux, I could believe you. I am really interested, but I am afraid of this becoming vaporware.

Ignoring the usual "vim", "emacs" etc., do try Sublime Text 2, if all you're after is an editor. So far it's proved unfailingly excellent for me :).

FWIW Chris stated in a previous post that Lighttable would be built on embedding CodeMirror, which, being web-based, should make creating cross-platform builds much easier.

Hear, hear on ST2. I don't have to do anything more than let people see me use it before they switch.

It's not just an editor. I use it for Python development. There's a plugin that uses Rope to give you completion, go to function definition, etc:


I've been using Emacs for months both on Linux and Windows and it runs much smoother on Linux, its' intended habitat.

Vim, emacs, sublime, Komodo, gedit, jedit... And that is just off the top of my head.

Are you saying that Sublime Text doesn't run well on Windows? I know it runs just fine on Mac and Linux.

Chiming in that it works wonderfully on Windows, although I had some problems with font rendering being ugly. Maybe it's just that I don't use ClearType, but I had to change some of the font options to get a result I'm happy with.

"font_face": "droid sans mono", "font_options": [ "subpixel_antialias" ],

Ah, okay. I was about to say originally that it is entirely cross-platform, then I realized there was some qualitative language in the parent post and wondered if maybe the Windows flavor was just flawed somehow.


Well, Sublime Text 1.x was only for Windows and was unique for rendering with DirectX, so one would hope it runs just fine there.

Vim and Emacs seem to work fine if you have Gnu's win32 tools.

You should try out Sublime Text. I have used it on all three platforms and it works pretty much the same on all of them.

Agreed. One reason I use it is BECAUSE it runs the same across all platforms, and is the slickest editor on all of them. I do development in multiple OS/environments daily.

>implying cross-platform gui toolkits don't exist

Since it will build on top of CodeMirror, I assume it will run on any OS that supports a modern browser.

I feel the part about it being open source is either misleading or your description is just really vague, even after the update. First you mention the core is open source, then you mention we need a license for which we can pay as much as we'd like. Now that last part doesn't seem very open source to me.

Fair enough if you want to make money with this, that's your right and I don't think anyone will try to stop you. But don't say it's open source when it's not.

Could you explain, if Light Table is released, what will I have to do to obtain a copy? If payment is involved, please explain to me how it is open source. What license will be used (MIT, GPL etc, if you roll your own, please outline the major parts)?

I was confused by this at first also. Open source means you can look at the source code. Sometimes it means more than that (often depending on who is using the term).

GNU's criticism of open source: http://www.gnu.org/philosophy/open-source-misses-the-point.h...

Chris: Just wanted to say "Good luck!"

Also, I saw you mentioned a patched Clojure compiler for column information. Did you use my patch? :-) Let me know how I can help with Clojure and ClojureScript contributions. I find working on the compilers to be a lot of fun.

This is no open source.

When I saw the Light Table demo, I wanted to build it for CoffeeScript, and I could do it.

This morning I found this, I got excited because I could use your interface and create a plugin for CoffeeScript. I was hoping I would be able to simply clone it and start working.

Instead I'm asked to pay X dollars to only get access to early beta?

> I'm a firm believer in open source

Yeah? Then make it open source. Develop it publicly on GitHub. Let people contribute. Don't make them pay.

> As such, I believe it only fair that the core of Light Table be open sourced once it is launched.

I believe I just gave you hundred dollars. Can I have the beta access now?

Developing things in the open can be a drag for a product first getting off the ground. People pull in different directions, want different things, and fork the project (and the community) in doing so. I'm a strong believer in free (as in speech) software, but the practicalities of doing so too soon in a product's lifecycle can kill it.

Personally, I'm more than happy for them to get the core done properly, and open-source later.

I just pledged $50. I did it for the license. As I did so, I realized: this must mean Light Table isn't going to be free software. Presumably I'm getting it at a discount for being an investor. Does it adequately compensate me for the risk of buying a nonexistent product? Shouldn't I get some percentage of its profits, if it's not going to be free software after an injection of more than $200K? Does the license give me lifetime upgrades, or only the first version?

These were my thoughts as I clicked through the Amazon payment buttons.

The problem is that it appears that $50 is not a discount on the final price at all, so you really get no added value by pledging money to help him make this a reality more than a year before you'll ever see anything.

He's now added a $15 license option, so I changed my pledge. He hasn't responded about the longevity of the license.

Yes, I've now pledged at the $50 level now that he added beta access to that.

$10,000 for a logo (and link?) from the front page of a popular project like this could be pretty cheap SEO.

I know Apache (the foundation, not the software) used to have a problem with sponsors who were basically buying links. Apparently sponsoring Apache was cheaper than trying to buy links on comparatively ranked sites.

It looks like there could be at least one company on http://www.apache.org/foundation/thanks.html at the moment that still falls into that category.

I am considering contributing but I don't have a sense of how long my vision needs to be. Do you have any sense of when a working alpha would ship?

Looking at the shopping cart [0] it looks like august 2012 for earliest testing, october 2012 for pre-beta and december 2012 for betas. All estimated dates, naturally.

0: http://www.kickstarter.com/projects/306316578/light-table/pl...

From the pledge section:

$50 gets you access in May 2013 $100 gets you access in December 2012 $200 gets you access in October 2012 $500 gets you access in August 2012 $1000+ gets you access in May

Ok, I will probably hold off on contributing then. I had alpha access for Sublime Text 2 for $60, and that seems reasonable to me. A $50 400 day pre-order is a little more than I can handle.

I signed up on Kickstarter just for Light Table. Now I can't wait until December!

I pledged $500. The only reason I'm being public about it is because I see people in here complaining about the cost of a license and wanting their dollars to translate into something more than this project simply becoming a reality.

I don't have a barrel full of money by any stretch of the imagination, but given all of the lispers and schemers on Hacker News, pg's essays, and so on, I find it surprising that there would be as much hesitation as there is about paying $50 for some of smalltalk's greatest hits coupled with an environment that lowers the barrier to entry for people to dig into the best modern lisp around.

I'm donating because I want to live in a better place, not because I want a license. The cost of not funding a project that could mainstream so many lost and forgotten great ideas while simultaneously promoting some of the best new ones, seems sort of like a crime against the craft of programming in general.

I mean no offense, just an observation.

Why is the first tangible reward a t-shirt and not the product?

If I was going to fund something, I would want to know my money is going to the product and not merchandise.

I don't think a YC pitch would go well if they said "And with the first $10k we're going to send out t-shirts to the first 1000 subscribers on the mailing list."

If it's written in Javascript based on code mirrorand the only thing closed source are a few plugins not related to the actual language plugins, then it should be fairly easy to roll a package with chromium or Firefox built around it. I expect a few show HN posts. I also hope for hosted solutions, and licensing for students. This seems like an excellent way to learn a language.

I think we will see done aspects turned into plugins and add ons for existing IDEs but that is not the same as building this intentionally from the ground up. If you can have function reference lookup, for example, and if it's built in Javascript, you can probably build in other forms of reference already on the Web. Pervasive passive information that you would normally have to search for.

Just backed and looking forward to trying this out!

I just have one question/suggestion just in case it makes sense. My main hesitation with using a browser-based IDE is that with native apps, I can use standard quick-keys to navigate documents very quickly using just the keyboard, such as flipping between documents, jumping my cursor to the beginning or end of a line/document and shift-arrowing/tabbing to make a selection. These are very basic and simple things, but without them, I would have to painfully use my trackpad/mouse to make selections, navigate the cursor around, or switch between documents. Just wondering if you're considering this type of functionality that will let me perform quick and simple actions through keyboard commands.

I'm a huge vim guy, and killed me to use the mouse as much as I did in the video. The real thing will be shortcuttable to your heart's content :) And all the standard things you would expect to work (like the ones you describe) will definitely be there before the end.

Was going to pledge the project, but saw "it is browser-based, it will packaged in a webview of some kind so that it feels like a normal app". Nothing wrong with browser-based editor, but I am comfortable with native app. That's my personal preference.

If closure is one of the first languages, how easy would it be to support java?

I am looking forward to Java support as well. I will be backing this project even if it isn't included, but the value in my workflow would be tackling Java.

$50 is way too much, especially since I don't use javascript or clojure and I don't see myself switching... ever. If there was C support or even python, then it would much much less of a risk for my money.

Python will come in at $300k, but one thing to keep in mind is that it is a highly extensible and completely open platform. There's no reason we have to be the ones to add support for all languages.

Just in case it's useful to anyone else, an HD version of the video (where the code is legible) is available on vimeo here: http://vimeo.com/40281991

T-shirt printing, packing, and delivery appears to be an unfilled niche.

I was really anticipating this Kickstarter campaign to start, but the pledge levels are a real downer. I don't mind paying $50 to get the licensee, but I should get beta access for that. This project looks really cool, but essentially preordering something over a year in advance (and lets face it, these timelines never actually get met) for the same price or more than you would pay when it is actually released without even getting beta access is somewhat insulting. What incentive do I have to pledge right now?

Where can I read more about the implementation details. I'm particularly interested in how well it will work on an "enterprise" app, e.g. 500K loc with some ugly dependencies.

Just a thought; when I read the title and started reading the description, I thought you were actually selling a giant drafting-table sized LCD.

If you wanted to integrate the IDE well with one particular piece of really nice kit, and then tell people that if they donate more than $BIGNUM, they get a nice workstation, all setup and integrated.

(just a silly idea. That's probably more of a distraction than you want. But it could be showy.)

Not trying to add one more "VIM IS BETTER" comment like happens on every thread about editors, but I have an actual, for-real question: who wants to help make a Vim workalike version of this? Because I would donate to that (in time, meager Vimscript skills, money) in a heartbeat.

I'd be very interested in pursuing this. I have some of this functionality already via numerous vim plugins, but not all of it. If I get a free moment in the next couple weeks I'll start writing something up about what already exists and what is missing.

Want to mention I think Light Table looks fantastic, but now that I have internalized vim's key bindings I am very hesitant to switch. That said, I still pledged on kickstarter.

Anyone have insight into how the tool that shows variables as it passed through functions would work in an extensible "port over the language of your choice" way? Is this reliant on stack tracing/debug tools or more rudimentary methods?

Very cool, but I had to ask the question: Is this similar to / also based around the Iguana translator? Thread here: http://news.ycombinator.com/item?id=3856159

If the guy is so excited about this concept, has he tried prototyping in as an (emacs/vim/eclipse/etc.) plugin? And if not, why? It seems that at least a few of those features could be shown off before he goes asking for $200K.

Yes, he has a working prototype, check the video.

I'd pledge the full $200K if they would take on C++ immediately. I am so sick and tired of IDEs that suck at delivering solid C++ coding environment. The best tool I've found for the job, to date, is Source Insight.

Thanks for making the change. I just pledged $15 for a license. I'll make good on my promise and spread the word around my networks, and hope others who pledged the least for a license payback in similar ways too.

All the best Chris.

I've thrown some of mine into the pot. As a Clojure apprentice, I'm eager to see the beta. However, I honestly doubt that some of the desired languages are even possible - C, really? Even JS seems dubious.

the javascript console in most browsers and JSLint already do this to some degree, Light Table just has to automate it. same goes for Ruby (IRB and ruby -S)

I was hoping for an open source project rather than a $200,000 product, but that's just me… Besides, Javascript being the main supported language is a put off to be honest. But there goes my $1.

mmm....if i contribute 10 bucks, i would expect at least a copy of the software. Plus, the UI is not too exciting and their expecation for investment might be abit high IMO

one of those times when continually clicking "refresh" is fun.

Woaha, that's a see-to-believe it. From $800 to $2 000+ in the time it took me to skim a couple of HN comments.

I am super excited, I have been doing backend development with rails and been kind of intimidated moving into doing more front-end stuff. I think LT is the answer.

So will this be a native app or will it need to be run inside the browser? I'm guessing the prototype was a web app but the final product will be native?

He has said it will run on top of the browser.

"In order to download the distributions, you'll need a license." Dosen't sound like a web app to me.

I understood it as a piece of software that runs in a browser, not necessarily a webapp. Could be local code that runs in your average browser, could be a stripped down browser that only runs this app, or it could be a native app that includes a webview as it's main interface.

So I guess we don't quite know yet. Will be interesting to see how he implements it! I just can't see myself using something browser-based as an ide but I suppose if it runs in an embedded "webview" that could be best of both worlds.

The FAQ on the kickstarter page says: "While it is browser-based, it will packaged in a webview of some kind so that it feels like a normal app."

I really hope they decide to move away from a browser based solution. I have personally toyed with the idea of making a web based IDE for front-end development, and had came across so many issues in my "experiments" that I decided it wasn't worth it.

Lets talk about editors..

CodeMirror/Ace as an editor.

While both Code Mirror and Ace ( A spin off from Mozilla's Bespin editor ) Have done a very good job, as Marijn Haverbeke ( Code Mirror author ) puts it, of "In-browser code editing made bearable". Neither are smooth / fluid / responsive enough for every/all day coding. If you read


You will get a good sense of why the browser is a very harsh environment for a code editor to adapt to.

Alternatively you can have a play with any one of the tools, that use Code Mirror and Ace.

Try to use them for an extended period of time and you will see what I mean.

    http://jsbin.com ( Code Mirror )
    http://c9.io ( Ace Editor )
I love the concept, and I'm going to be making a donation either way. I just hope that they move away from the idea of a browser based editor. Before investing a huge amount of time and energy into it, and end up with an inferior product, largely attributed to the platform it's running on.

As a front-end developer, I'm all for making as much as I can browser based / cross platform, and pushing the limits of the browser. But from personal experience the browser isn't the place to be writing the code to do it.

IMO it's a shame they stopped development on Bespin. It felt so smooth and fluid to code on because it used canvas. You totally forgot you were coding in a browser. Ace is great but it just doesn't feel the same eh.

For a change, I wish this wasn't web-based, but based on Chromium and really fast.

(And with a great plugin system).

It is all of those things :)

May 2013? I can't wait for so long.

Will it have any support for popular JS modularisation libraries like require.js?

$200k seems like an awful lot to me. I don't see why you need a massive team to make an IDE. It's not like you're making a game, where you need artists, programmers, composers, etc. All this is is programming.

so the software is opensource, but they are providing official binaries for a Fee. Hmm..very similar to what XChat did a while ago

With the new early beta access for $50, I'm in.

I don't get it. This proposal sounds like Eclipse that exists today plus a couple of small new plugin for the more aspirational bits.

The current pricing structure is discouraging backers, in my opinion. You tell us that when it's released it will most likely be pay-what-you-want, but if we pay early and support the project we are pushed into $50. I can guarantee you that if you do pay-what-you-want, your average isn't going to be $50.

I don't know what percentage of backers are the kind that do it merely to "support" a project, but I would guess it's not more than 10-15% of all backers for a given project.

The rest want some value for their money beyond feeling good about backing. You state you chose $50 for a license because that is what you think it is worth. I think you'll find out that what you think people will pay you for something and what they are willing to pay differs by a factor of 2-3.

So most of the people that want what you are going to make, even a little, probably don't want to spend more than $20 or $25. The difference between that and $50 is made up from goodwill, but that's really asking a lot of people.

I would recommend making some more tiers between $5 and $50, where everything above $20 gets a license. This way you capture the people that just want the product, and give others the ability to show their goodwill as well.

This isn't the kind of project that really needs swag. Nobody really wants a t-shirt. If you do something like promise to list backers (and their associated tiers) in the "credits" of the app, that would probably be great for most people.

edit: typos, and wording

Agreed. $50 for a license it too high. Lowering it to $15 for a license would put it in impulse buy territory for most people, and you'd likely get a lot more backers and receive more money. Personally, I don't know you at all, so I have no reason to trust that you will produce something good, and so I don't want to gamble $50 on it. But, I would gamble $15 on it.

The other software kickstarters I've backed have included a copy of the product for the lowest tier (usually $15), and then added swag like t-shirts at the higher tiers.

Something like this:

$15: Light Table license

$30: Light Table license and included in the list of contributors

$50: Light Table license, included in list of contributors, and early access to beta

$100: Light Table license, included in list of contributors, early access to betas, and t-shirt

Lowering the early access will get you more beta testers, and raising the t-shirt higher will make you have to ship less t-shirts.

I couldn't change them even if I wanted to now that money is involved. But I think looking at this as buying something is also inaccurate. Ultimately kickstarter is a way of helping things turn into a reality - not a place to shop. It's basically a form of donation, that just happens to have some reward associated to it.

"Ultimately kickstarter is a way of helping things turn into a reality - not a place to shop."

I disagree, I think shopping is exactly what Kickstarter becomes. You can use the platform for a lot of different realities. But, really, I didn't back the pebble watch because I thought the guys were nice, or because I thought the world needed another watch, or another thing to charge at night. I backed it because I like having new shit, I wouldn't of done so if I didn't get something physical from the money. As much as kickstarter is the poormans way to vc projects. VC's want to see something for their money too.

Even if my only reason for backing projects was to see projects become a reality. I'd still be shopping in a way, shopping to find projects I wanted to push into a tangible place. But even at that point, I still think it's the minority.

Donation based projects are more than fine, i'd love to see light table exist. But don't underestimate affordable donations, or fair rewards.


Kickstarter isn't "supposed" to be an e-commerce site, but that's what the majority of its users seem to be using it as. That's the emergent use case that's become almost a standard for the site.

Regardless of how we might feel about it, pricing needs to accommodate this user behavior. Not the other way around. Expecting users to change how they use Kickstarter is a recipe for sub-optimal subscription, if not failure.

"That's the emergent use case that's become almost a standard for the site."

For the consumer electronics. I'd say it's absolutely about crowdfunding and supporting the creator for artist and artwork projects, which are the majority that I back, as well as many people I know.

Precisely -- I have argued before that crowdfunding is not an option for a startup or any other commercial activity. But if we see it as selling in advance, than it makes all the sense, and pricing and everything should be set up just as if they were selling an existing product.

However, if Kickstarter doesn't allow changes in pricing once the project is set, it might not be the best option. Why not simply set up a personal pre-selling page, accepting payments via PayPal or any other service?

"if Kickstarter doesn't allow changes in pricing once the project is set, it might not be the best option."

Can not change current reward level's, at least not after they have a purchase. But you can always add more levels. So as much as that does not alleviate the fixed prices, I think if the prices are well thought out, then more reward levels were added based on demand of the market, at that point I think it becomes a pretty successful platform for a startup or commercial activities. Although there are definitely exceptions.

"Why not simply set up a personal pre-selling page, accepting payments via PayPal or any other service?"

Awareness. If I started a new company that sold shoes, set up a seperate page. I would not have the instant credibility or the traffic that kickstarter has. As much as there is a small chance if the market demanded what I had, that I could push it viral without kickstarter. I do believe that the kickstarter platform has proved itself over the last few months. Multiple projects being pushed over the million dollar barrier, projects spanning from pure hardware to pure software. I think there are plenty of cases where kickstarter should be a last option. You're still preselling product, instead of equity, all while gaining awareness/free publicity while not taking very much risk.

On top of that, with not using kickstarter, you're losing not only the traffic, but the impulse buys. At least a lot of them. At the discounted prices, people jump on bandwagons. For instance, if I see something that has taken off, I will in a lot of cases "back" the project, even projects where I didn't think I wanted or needed before logging on to kickstarter.

Can't you add new tiers at any time? You could make parallel $15, $30, $50 and $100 tiers. Also, nobody has actually paid yet, since the money isn't collected until the end of the period. People can change their amount and reward selection at any time before the backing period ends.

I agree with you in theory that kickstarter isn't a place to shop, but in practice it appears that offering 0 marginal cost rewards (e.g. software, videos, early access, media, etc...) at low tiers is a very successful strategy.

That doesn't mean you shouldn't give actual value to the people that are making this possible for you. People that donate the full license price over a year in advance without even knowing if the end result will be realized should at least get access to the beta. The people that pledged $100 can still get earlier beta access, but not giving any kind of early access at the $50 level is a deal breaker for me, and I know I'm not alone.

That's not a safe assumption to make of Kickstarter. If the intent is really to treat everything as donations then every tier should include "acknowledgment as a backer". As it stands today it's only stated for $1000+ level and that's not really acceptable if you consider everything a form of donation. I saw very little in the description or "rewards" that really treated backers as genuine donations rather than just simple pre-orders (that includes access to betas and/or user testing). It was rather confusing compared to nearly every other project I've contributed to on Kickstarter.

You're right that the tiers can't change much now, but the way it was structured and worded is indeed discouraging. I know I was certainly discouraged as a result, and I'm currently on the fence as to whether to contribute now or wait and see how it turns out. If this project doesn't succeed in making it's goal I would really recommend looking at the more successful software projects in detail and notice how they go out of their way to message and recognize backers in a way that makes them feel like their genuinely contributing, even if only in spirit, and orient their rewards accordingly. Adding buttons, logo stickers, t-shirts, posters, etc. are the kind of things that a community of supporters will desire and from what I've seen so far, go a long way in encouraging larger donations.

edit: removing suggestion to rollup rewards due to their potential for complicating things.

You could easily change it if you want to, send out an update and everyone who's contributed will be notified of the changes.

I must admit that I'm having a really hard time justifying donating to this project, if it really will be pay-what-you-want when it's released.

Looking at the kickstarter page, it looks like you're completely missing out on the low tier donations that many other kickstarter projects get the most money from

I love how the top-voted comments on HN are always "the price is too high!" And yet, Light Table has already raised $22k in 4 hours-not a bad start.

Yeah, but it isn't even remotely enough and the majority of what has already been pledged are from the hype wave of HN that will disappear in a day or two. You only get the hype wave once.

I actually think the amount would be much higher had the $50 included beta.

What data are you basing this off of? Have you ever tried selling products at significantly different prices and measuring revenue/profit? I have-as one example, I've sold the same PDFs at $.99, $9.99, and $49.99. The $.99 and $9.99 price points got about the same revenue, with $9.99 getting much higher profit due to transaction costs. But $49.99 got much, much more revenue and profit than either of the lower-priced options. And I've seen this over and over again, with a variety of products.

What you said is confirmation of my guessing. I'm saying if $50 includes beta (currently only $100 [EDIT: just saw $50 for beta added, thanks for listening to HNers ibdnox! Oh, wait, that access is much later than the $100 price point..]), then the amount he may be getting may be higher than the current amount.

FYI, no data to base off, just guessing, hence the words "think", "would" in my post you're replying to.

Of course it isn't a bad start. But, as others pointed out, the price point is pretty high as it is merely a bet on the product.

I myself would have considered buying/donating, although I wouldn't use the first version (not doing any Javascript or Clojure projects). But I see value in what is planned here and would be willing to bet twenty or even thirty dollars on its success. It would be a bet that the final product, which I basically bought, is good enough that people implement plugins for it that I could use (ruby/php/java/actionscript/etc).

But $50 is a pretty high bet in this case, and I would bet that I am not the only one in the described situation...

You get a lot of passion at the moment the bargain goes down.

No, it's not a form of charity. Kickstarter even goes so far to make it clear they aren't in that game. In fact, they even say this:

"Kickstarter isn’t charity: we champion exchanges that are a mix of commerce and patronage, and the numbers bear this out."

You've hit the nail on the head. I'll back for $15, but not for $50.

Kickstarters are a bet: I expect 50% of them to succeed, so I have to be able to extract double the backing value per project for it to be worth backing.

I'm obviously not the only one, which is why the $15 price point is so effective..

That would be absolutely perfect. Plus, let's face it, this won't be a success in the long run if a lot of people aren't using it and there isn't a community behind it. The goal right now needs to build a large user base, not squeeze every last dime you can get in the fundraising portion. This is ultimately supposed to be a platform right? The profits will come from the long tail with upgrades to a large number of users and add-ons like new language support, etc.

According to me tools have practically zero chances of survival without communities rallying behind them.

This looks like such a cool idea. I think their primary concern must be to release a basic thing out first and then iterate over it. This way they will get both contributions and the product going.

Decide what your design goals are, define what code and quality goes in and out of your repo. Build something basic, release it and then iterate. Contributions and donations will follow. But if you are expecting mass adoption for a tool designed primarily for Clojure then, Clojure itself is in low adoption stage now. On top of that people who use clojure wanting to use will be fewer(Emacs + Slime really works well). So the initial payment model with only Clojure support may not bring them much money. This definitely has to be a platform like Emacs. And then the money can come from upgrades, major revisions.

Also it would have been nice if language used to extend this was Clojure itself.

We desperately need some modern GUI candy on an Emacs like editor which can be extensible by Lisp.

Out of curiosity, could you give an example of the software kickstarters you're talking about?

While in an ideal world, things like this could be free, in reality, the amount of time, effort, and people involved to build it in the time frame we're all wanting this means they need money and I'm not sure having rewards for $15 is going to accomplish that. That's the unfortunate reality of what is wanted, and what it takes to actually get there. :\

I'm talking about the recent PC game kickstarters that have been wildly successful such as: http://www.kickstarter.com/projects/inxile/wasteland-2

I know that a game and IDE are not directly comparable, but they do have some things in common: 0 marginal cost and expensive to implement due to skilled labor requirement.

Textmate is 39 Euros or approximately $53.

Yes, and it is a trusted piece of software that has been around for ages and I can start using it right now.

Textmate is certainly cool. Then again, I've tweaked my Gedit to work like Textmate (and it's Free-as-in-Beer, and I can use it right now).

However, neither do what Light Table does. I'm also certain that there will be plugins for practically every language with a REPL, given enough time and the fact that it's open source.

I agree. I was eager to contribute but saw the page and didn't do anything. The rewards just seemed kind of stingy and the target seemed high. It's software, zero replication cost. Everyone who contributes should get a copy, and additional rewards should encourage additional generosity.

$200k is presumptuous, really. If you look at the top-funded technology projects, they all have low goals and excessive results: http://kck.st/J7kUqn

A high goal on Kickstarter /is like a high price/ and seems to trigger similar supporter/customer behavior.

When I saw the Light Table video on HN I flipped out. I want this! But when I saw this campaign something cooled my jets. I'm not sure I'd pay $50 if it was for sale right now. If I'm supporting its creation, and speculating at that, I need a more alluring price and/or a story that makes me forget about all price concerns.

If I were the creator, I'd seriously consider pay-what-you-want.

I would argue few are as ambitious either. To put this into perspective, the VS team is numbered in the thousands. We're talking a small team, but a team nonetheless, and it's hard to imagine being able to keep a team together for a year at less than 200k.

In this case, it's not a matter of presumption, it's a matter of knowing the reality of what is involved in actually building this thing.

If that's what it takes, by all means they should get it. I think the best advice then is to simply make the offer as remarkable as the ambition, so we all get on board.

Edit: I did't mean to suggest the goal is unreasonable or unattainable. I was just trying to explain the not-positive reaction I had to seeing it. Like shopping, not all price reactions are logical.

Similar experience.

If it's worth $50, give it to early-adopters/supporters for 50%, because unlike buying something for $50, they don't get it now (and there's many a ways for a project to not deliver). In this promotion phase, it's good to have exciting specials. (e.g. Notch's 50% beta Minecraft) Marginal cost of software is zero, more users = more evangelists etc.

Saying "license" looks non-free. I guess it includes the closed-source plug-ins (?). Good to clarify.

And they doing the opposite of this because they ask for double the price for early access, but currently it's the most popular option with 91 backers.

Dead horse but, without being able to see upvotes, Chris has no way of knowing if 2 people agree with this or 200. This is a wasted opportunity to help.

For what it's worth, I agree, especially on the t-shirt. I stopped reading after $20 because (a) I sure as heck don't want a tshirt and figured everything else would include it, but that doesn't matter because (b) I have no desire to "invest" more.

For awhile now though, there's been a $15 tier, and currently only 3 backers have chosen that option. Given the cost ineffectiveness of a t-shirt, particularly at a low price like $15, other than Chris' eternal gratitude, what sort of recompense would people want for that level of support? Almost 30 times the number of people went for the $100 beta access level, and only slightly fewer went for the $50 license, so apparently it's popular enough, but what's the best way to get support at a $10 or $15 level?

>but what's the best way to get support at a $10 or $15 level?

That's obvious: Include a full copy of the product at that level. Call it "no support" or something, but if you're going to be doing pay-what-you-want, then go ahead and start now.

He can obviously charge what he wants, but I think my parent was spot on. $50 is too much as the first tier where you actually get something [you want]. This is especially true considering the final product will be pay-what-you-want.

I think playing with support options or possibly upgrade paths ($20 get you free upgrades for 1 year, $50 gets you free upgrades for 2 years or somehting) might be a solution...though I'm not sure myself.

As of right now the $15 option has 0 backers. There is really no benefit to donate below $30 right now. I don't see why giving everyone beta access would be an issue. It gives some value to the lower tiers. Invite priority could still be from greatest to least but that last tier would see something eventually.

Being able to only pay the difference when buying a licence is another option.

I think its always hard for software kickstarters to offer anything interesting besides the actual product. As you've said, in this case, even merchandise doesn't really make sense.

The only tier, apart from a 15$ software tier(as its been standard), that would make sense to me personally(and financially) would be an option to vote on what language he creates support for first(after he's done with the initial Clojure version).

I agree. $50 is too much for a prototype. But I would gladly pay $50 if this is already a developed product and ready to be used.

The main thing holding me back is the supported languages. I don't use Clojure and I'm not a big enough Javascript fan. Suggestion: Add more languages if you reach enough backing

$200k: JS and Clojure

$250k: JS, Clojure and C

$300k: JS, Clojure, C and ...

I don't use JS much or Clojure at all. If I were the developer, I would look at adding the top used languages post haste: C/C++, Java, Python, Perl, etc.

It seems like it might be particularly well suited for HTML5/CSS3 development as well, since it runs in a browser, and that would be great to have.


C development in this manner isn't a particularly trivial matter. C header compilation and decleration alone complicates this.

libclang could help a lot here, as it gives access to the internals of a C/C++ compiler all the way from tokenizing to code generation. It also has built-in support for all kinds of IDE features such as code completion and fixit hints.

and thus my INCREDIBLE description

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact