Stypi (YC S11) Is Etherpad Reborn | Hacker News Reader

Stypi (YC S11) Is Etherpad Reborn

techcrunch.com

138 pointscanistr72 comments

Not to be a dick, but who cares? This just seems like a lame proprietary rip-off.

Etherpad is alive and well, and it's totally F/OSS - Etherpad Lite was just released and it's SUPER easy to embed in your own apps: https://github.com/Pita/etherpad-lite

Easy, Because it's a YC company. and everyone knows Paul Graham farts unicorns.

To get real for a minute. They may be able to make this into a nice money making niche product. There's plenty of open source products that people still pay money for. Just look at OSS content management systems.

From the install docs for etherpad-lite on Github, step one:

As root:

Okay, two words and we've just lost 98% of the computer users in the world.

Install all dependencies. We need the sqlite develob libraries, gzip, git, curl, wget, libssl develop libraries and python:

... and now we've lost most of the remaining 2%.

"SUPER easy" for you, "SUPER easy" for me, but the other 99.98% of the potential customers would also love to use the product.

Seems to happen to a lot of products that are 'simple'. It happened to Dropbox ("Why would I use this when I can do these other twelve steps to do nowhere near the same thing") and the Ipod ("Why would I need this when I have a CD Burner and a Discman" (srsly)). I think it's safe to ignore claims like it :)

Two reasons to care:

1) This is a YC company after a couple months. They'll continue to refine the product; imagine if Facebook had gotten set in stone back in 2007, released code, and didn't themselves continue to develop. I love F/OSS for some things, but this isn't the kind of software which usually gets developed well entirely by the OSS community -- it's design/UI/UX heavy.

Even if it were just a 100% clone today (which it isn't, it's better and worse in different ways), I'm excited what it will be in 6mo.

2) Social community and network effects are a whole lot more likely for this than for an embedded app. Again, standalone Facebook is way less interesting than facebook.com.

right, so let's all praise it and talk about it in a few months when it's a good product :)

Etherpad's well suited for editing human-speak documents, but not a good fit for collaborative code editing. This is more in the vein of SubEthaEdit on the web, in vim, and in other editors (I presume).

This is going to be a friggin' boon for phone interviews.

Thanks for the link to Etherpad Lite - looks like a good project. That said, I set up Etherpad a long while ago for family and a few invited friends and installing it was fairly trivial. BTW, very roughly similar, but: Wave in a Box is also an easy install and I like the slightly simpler interface (compared to Wave itself).

Linux is an F/OSS Unix, but there was still room for Mac OSX. Some of us like nice things maintained by somebody else.

To beat Etherpad and Google Docs:

Let me hit ctrl-I while typing, type something and have it appear in italics, and then hit ctrl-I again to turn off italics. Likewise for bolding, though I use less of that.

Let me apply URLs.

Do this (while keeping it fast and without introducing any bugs worse than Google Docs sometimes crashing and IEtherpad sometimes going unresponsive) and I will switch.

I wouldn't even need ^i and ^b. I wish they added a Markdown "syntax" mode. And it could be interesting if you could download the history as a git repo.

Just used this for a phone interview. I knew it was a bit risky to do that on launch day, but decent support for syntax highlighting, indentation, and showing the other user's cursor made me think this was a use case this app was built for.

It was great for 30 mins but after that we started to get connectivity issues and had to move to TitanPad. Once it's more stable I'll switch for sure though.

Does YC invest in innovation? If so, why am I not seeing innovative startups?

If not, is it because iterative 'also ran' stuff is a safer investment?

I'd say some of their examples have resulted in innovation (Heroku, Airbnb), but overall, I think most incubators seem to play it safe. I really wonder if the next Google/Facebook sized company will come through an incubator or sidestep that process altogether.

Innovation in the form of incrementally improving existing ideas, or providing a better implementation of a current idea, is still valuable to society and to investors. Google was an "also ran," as was the iPod. Both products (search and MP3 players) had been done before.

Cool. I was looking into Cloud9 yesterday, but Stypi really does the rick, due to the syntax highlight and collaboration (cloud9 did not work for me), non signup required and ease of sharing (simple url).

Great job guys!

Looks like they're using Ace underneath too. (http://ace.ajax.org/)

where is the vim plugin?

I was wondering about the same thing.

Feature request: please add Objective-C to the list of supported languages. Thanks!

I'm super psyched about these guys - what they're working on has all kinds of potential that ether pad never had. It's one of the missing kinks in the modern development toolkit. I hope we'll get a chance to look into the API soon - I know they're working hard, but I don't want to be patient :) we have visions to fulfill!

Yes! I want an API!

Looking forward to trying this! There's a lot left to do in collaboration, and a commercially-backed piece of software will be able to achieve some remarkable things.

Collaborative code editing, in particular, is really an unsolved problem, so I can't wait to see where this goes.

Feel the collaboration! stypi.com/pi2o36wr

Congrats on the launch Stypi team. Just curious about the underlying tech. Are you using operational transformation (OT) like google docs does OR are you using an alternative such as causal trees (CT)?

The article notes they didn't use any of EtherPad code. Why not?

Probably to avoid the GPL?

EtherPad is under the Apache License (non-viral, like BSD)

EtherPad is Java. That is one of the reasons I never installed it on my own server.

Looks like node + socket.io for the push updates. Very snappy.

Looks like a cool web app, although I fail to see this as a commercial product. I would use this only as a free product, and I'm sure as hell that my friends won't pay for this either.

Maybe I just lack vision, but I don't see how this app can possibly pull in big revenue. Can someone please prove me wrong?

Businesses definitely need apps like this. Beats the hell out of emails and MS Word.

I'm not sure you understand what businesses actually use email and Word for.

If it added a few more features (like an API, cloud storage and embeddability) I would pay for it.

Im surprised they don't have markdown or textile in their list of languages.

Suggestion: The style should stay user-independent, but the text syntax should be shared. I mean... it would be bizarre if one person thinks that they're editing Python and another is editing Java

This is great. I'd love to see pair programming accomplished somehow with this (on the web, not via screen or buffer sharing). Can we integrate a compiler or static type checker into it?

I loved Etherpad and am excited to read this. Are you guys using Ace for the editor?

So their competitors are just

    Google docs
    PiratePad
    collabedit.com
    primarypad.com
    TypeWith.me
    Sync.in
    EtherPad Foundation
    iEtherPad.com
    titanpad.com
    etherpad.tugraz.at
    etherpad.mozilla.org:9000
    sketchpad.cc
    okfnpad.org
    pad.telecomix.org
    twiddla.com
    ...

Now the real question that I'd love to be be answered is - why would anybody fund this?

Because if Stypi can become one of the top 4 companies on that list, the VC firm gets a stake in whatever Stypi decides to build with the huge number of users that implies?

Not to put too fine a point on it, but, it's easy to see why you wouldn't invest in it: because you can't afford to put that much money behind a 1:10 or 1:15 risk. But that's exactly the problem every VC firm in the world has: finding the right set of 1:10 bets to put their money behind.

I feel like, if you're going to complain about people funding stuff like this, you're really complaining about the whole VC model. And sure, I don't like it either. So go build a company without it. Plenty of us here have or are doing exactly that.

Because if Stypi can become one of the top 4 companies on that list...

The question is why do you or anyone else believe that this is a distinct possibility -- i.e. given this team, this product and this vision, why does someone with money believe that this represents a good bet?

I agree with your assessment of the VC's challenge. However, VCs don't put money into every long bet that comes their way. The different between a 1:2 bet and a 1:10 bet is obvious -- distinguishing a 1:10 bet and a 1:10000 bet is where the art comes in. (That is distinguishing a long but feasible proposition from bullshit.)

Sure. The point is just the naivete implied by being surprised that someone would invest in an Etherpad clone, as if anyone was likely to have simply invested in an Etherpad clone.

I see people all the time on HN who think that the logic of a situation inescapably reduces to a single, simple statement or set of axioms which coincidentally agree with their own viewpoint. They are genuinely shocked and bewildered that a large segment of the population would fail to reach the same, simple-to-reach conclusion, and from this they conclude that everyone who does not arrive at that conclusion is defective.

Even if 50%+ of the opinion-holding population disagrees.

Because they are unmitigated narcissists, it never enters their mind that perhaps there are factors unseen to them that make the analysis more complicated for those other people.

I think this is related to what Nassim Taleb calls "epistemological arrogance" -- refusing to question the validity of one's own knowledge, despite significant evidence that one's knowledge may be insufficient or invalid, e.g. 50%+ of the population has a different opinion, or people with a compelling need for high-quality knowledge (VCs, in this case) are arriving at different conclusions.

For concrete examples of this, see the comments on HN article which involves a discussion of race or gender relations.

You do raise the important point that one should never assume one knows everything, but 50%+ of the opinion-holding population is wrong about a very large set of ideas (though which 50%+ varies from idea to idea). As such, the fact alone that the majority of a population disagrees with an idea should carry little weight when deciding whether to keep or abandon an idea.

Agreed. However, that fact should give you pause, and if you reject the majority opinion (or merely a plurality, or just widely-held) you should be able to articulate a solid reason as to why it deserves rejection.

That's a bold claim. Even if it becomes one of the top 4 companies, what's the business model? Most of the sites I listed are non-commercial, absolutely free. It's a bitch to compete with free services, and pretty much identical functionality.

You must also realize that the sites I listed is just a small percentage, something I googled in 2 minutes. There are many many players in the collaborative editing "market".

How could you possibly think you know this company's investment pitch based on the information we now have? You just want to bag on YC companies. And I don't care, except that you're doing it pretentiously.

I actually do love many YC companies (reddit, justin.tv, disqus, airbnb, optimizely), but not this one.

Or the last one you commented about, which you also vacuously slagged.

Both snapjoy and stypi seem to lack innovation, at least from the information they are giving us. Take hipmunk, for instance, that's a great example of a fresh idea.

One of my coworkers here at Google wrote a webcrawler and search engine - probably the first webcrawler, according to Wikipedia - back in 1993. It was an academic project, deployed completely for free, with no possible way to make money.

That didn't stop Google from making hundreds of billions of dollars off search 15 years later.

When investors look at a company, they look at what it could become, not what it is now. Most gigantic megacorps started as tiny niches that nobody ever believed they could make money from.

Google didn't succeed because of a crawler. They succeeded because of pagerank algorithm, a new invention (plus hiring the smartest people they could find).

And I'll ask again - what's so groundbreaking about this project?

You're missing the point which is: when Google was funded, it wasn't Google, the king of search, as we know it today, but yet another search engine in a saturated market, started by 2 academics with no business background.

The reason they were founded was that they convinced the investors that their plans and ambitions go beyond what they've built so far.

It's a remarkable lack of imagination to not see a potential in a good, collaborative web editor. Sure, if they don't ever write a single line of code, it's not going to be anything special. But they probably will and they probably have many ideas on how to move this project forward.

An obvious evolution would be to turn it into a web-based IDE, competing with the likes of cloud9.

Or they could make it an embeddable editor widget usable by others (it's surprisingly difficult to build a decent web-based editor and a lot of web-based software (a forum, a commenting system, q/a sites) needs a good editor).

Or many, many other possibilities. Stop thinking of it as a finished product. Think of it as a first step in a thousand-step journey and try to imagine what the other 999 steps could be.

It's you who is missing the point. Google was not "yet another search engine", they invented something new, which gave them the edge.

You keep talking about the future possibilities, but you fail to point out what gives this particular project that edge. What is it?

FWIW, PageRank is a relatively minor component of the ranking algorithm (I dunno why people latch onto it...maybe because it's a public, patented, easy soundbite), and the core Google ranking algorithm was entirely rewritten by Amit Singhal in 2001. The ranking edge that most people associate with Google wasn't even written when Google got its funding, and its inventor hadn't yet been hired.

I don't follow. Wasn't Larry Page the inventor of PageRank? And didn't he make it while still at Stanford?

I'm saying that PageRank isn't the ranking edge that Google is famous for, and that if you applied the ranking algorithm as invented by Larry and Sergey to the web (even the web at the time, before everyone started to game Google), it would suck. It might suck a bit less than AltaVista, which was why they got funded, but it's not at all the quality that people have come to associate with Google.

Ah. When you say 'the edge Google is famous for' I think of PageRank.

Exactly.

I think folks enjoy thinking that something they can also understand (like PageRank) was the reason Google was successful...

The real question I have is: why posting a mean question without making a slightest effort to answer it yourself? Did you read the article? Did you try the product yourself (or all of the competitors you've mentioned)?

I, for one, believe that collaborative editing should have been added to every authoring application, not just editors, 10 years ago. I still use shared screen sessions for that.

I also thought exactly this until I read "vim plugin".

Clearly this is just a start, a way to have something running and make people talk about it, a way have real users testing their technology... And being a YC startup help having a relatively large audience even for this kind of launching with an uninteresting product, so the real world technology testing will work, and I'm pretty sure that's all they care about for now.

When they start developping plugins for other apps using the same already-tested-and-approved techno, it will get interesting. This seems to me like a case of "launch now, code later", which in this context may be a good idea.

Oh, and by the way, I'd like an Emacs Lisp implementation and an open protocol (and a unicorn).

Simperium is another competitor (and a YC alum, as was AppJet).

As everyone else replying to you has noted, these guys have grander ambitions. So does Simperium — they aren’t just doing Simplenote (see http://simperium.com/). (Full disclosure: I contracted with them in the past)

Because anybody who works remotely knows what a titanic, epic, massive, planet-sized pain in the ass it is to do remote code editing.

And this product does it out of the box. I've worked for four years remotely, and with OSS projects, and this has always been a niche that needed to be filled. I've tried subethaedit (requires someone to have open ports to host), coda (which uses subethaedit's libs under the hood), vim + screen (requires having a server + logins), etherpad (no code highlighting), and git+tight sync times, or just ever more towering pages of gists.

Having a websocket based collaborative code editor with line numbers and syntax highlighting is 95% of the way there, where as everything else is at best 75% there.

So, yeah. I would pay money for this.

Have you tried using webmeeting software for this ( like webex, or gotomeeting, or even skype )? Then you can just share your whole screen / environment with someone remotely. That is what I usually do.

Actually yes, i left skype off the list accidentally.

I've successfully pair programmed via skype screensharing, but the image compression can often be a problem (as well as refresh rate), and there's no way to actually share work/alternate control in a pair.

It sort of works, but gets like a 65-70% score from me. Can get the job done, but seriously inhibits productivity due to lack of features.

You could have said the same thing about Dropbox when they launched. Or Google.

Lots of competitors is a sign there's something here that people want. Unless the incumbents are protected by some barrier to entry, the winner will be whoever executes best, even if they are a late arrival.

Not quite. Both Google and Dropbox brought immense improvements to the state of the art of both of their respective niches.

I have a hard time at the moment trying to figure out what benefits this has over Etherpad clones or Google Docs. I'm sure there's a use case that they're trying to tackle, but it's not obvious to me.

In the article, it's mentioned that they are building a platform whereby native apps can add collaboration features, and they mention Photoshop as a possibility for that. That would be huge.

You're not funding position and you're probably not funding velocity, you're funding your estimate of acceleration. At t = not much, what they have built in aggregate is a poor predictor of future value.

Your criticism reminds me strongly of patio11's prediction:

"What's this?! You dare call yourself a startup? You're attempting to solve a problem which other people have attempted to solve before, and your solution looks like something two guys could have made in three months! Why are you in YC and not me?!"

http://news.ycombinator.com/item?id=2862853

I am not involved with Stypi but I think the key points from the article are "it wants to let native applications like Photoshop sync changes between multiple users" and that "they’ve laid the groundwork to let other applications sync".

In other words a text editor is one instance of an application that can use the underlying service. Think about that for a second. The concept of collaborative editing as a service that can be adapted to ANY application is HUGE. For example, an engineering CAD app would be more powerful with this feature.

Going further,if you consider that machines can also "collaborate" then it get's even more interesting. An example of human/machine collaboration could be a real-time reporting application.

Reply on news.ycombinator.com