LWN: Comments on "EtherPad source code is free, now what?" https://lwn.net/Articles/367879/ This is a special feed containing comments posted to the individual LWN article titled "EtherPad source code is free, now what?". en-us Fri, 03 Oct 2025 05:49:09 +0000 Fri, 03 Oct 2025 05:49:09 +0000 https://www.rssboard.org/rss-specification lwn@lwn.net EtherPad source code is free, now what? https://lwn.net/Articles/370602/ https://lwn.net/Articles/370602/ johnmclear <div class="FormattedComment"> Edu has its own host etherpad service, schools buy a hosted pad for their <br> school or district (see primarypad), also see ietherpad for another consumer <br> option.<br> <p> Etherpad is crazy heavy on resources and isn't mature enough yet in the open <br> source space. Most decent info is available here<br> <a rel="nofollow" href="http://groups.google.com/group/etherpad-open-source-discuss">http://groups.google.com/group/etherpad-open-source-discuss</a><br> </div> Thu, 21 Jan 2010 15:59:03 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/370040/ https://lwn.net/Articles/370040/ pkern <blockquote>Financial records, for example, might be convenient to access from multiple locations, but a hypothetical "Gnucash Online" service would not need to share information between users concurrently.</blockquote> It seems to be that someone did not yet use Gnucash in a group. The locking mechanism of Gnucash Offline requires a fair deal of coordination between multiple users (which we have in the non-profit organisation I work in, using Gnucash). Concurrent updates was something we envisioned for a Gnucash replacement. Sadly the task of creating a new featureful application was too large for the NPO at hand. Sat, 16 Jan 2010 10:33:12 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/369051/ https://lwn.net/Articles/369051/ giraffedata <blockquote> Normally when a company buys another company they get the whole shebang - including the liabilities. </blockquote> <p> I'm saying there is no legal concept of buying a company. You buy individual pieces of property, and a "company" isn't one. There are lots of kinds of transactions that we informally call "buying the company," and the most common is where you buy all of the assets of the company and leave the liabilities alone. That seems to me to be the natural structure for the Google-Etherpad deal. <p> <blockquote> Otherwise it becomes a quick way to void contracts </blockquote> <p> One of the reasons corporations exist is to give people a way to void contracts. People who contract with corporations know this, of course (it's one reason corporations are required to have words such as "corporation" in their name) and usually require a natural person to guarantee performance of a contract with a corporation (I have personally given such guarantees many times). (Google-size corporations can negotiate on their own credit, but they are the minority). <p> <blockquote> Also, companies are bought on a routine basis by buying all the shares - it's one of the common ways of doing a buyout. </blockquote> <p> Yes, as the post you're responding to says. It also says this is an uncommon way to buy a company, because of the liability issue. (People aren't really worried about the outstanding contracts as in the Etherpad issue -- the liabilities no one knows about yet are the important ones). <p> <blockquote> You bid x per share, </blockquote> <p> That doesn't mean you're offering to purchase the shares. It's also the best way to state the purchase price of a publicly traded company's assets, because that's how much cash the shareholders get when their stock is ultimately cancelled. <p> I guess I should acknowledge now that there are special areas where the law does recognize the concept of buying the whole shebang -- they go under the terms "bulk sale" and "successor liability." But they're details that aren't worth discussing here. Also, directors of a corporation are responsible for operating it in a way that it can be expected to pay its debts, so in the case of a flat-out distribute-the-assets-and-screw-the-creditor transaction, the directors have to pay those creditors and a creditor can even get the money back from the shareholders. But that's nothing like making the guy who bought the company's assets pay those debts. Thu, 07 Jan 2010 17:03:19 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/368952/ https://lwn.net/Articles/368952/ jjs <div class="FormattedComment"> Remember, Google didn't buy a product line off etherpad, they bought the whole thing. Normally when a company buys another company they get the whole shebang - including the liabilities. It's done all the time, btw. Otherwise it becomes a quick way to void contracts - which would be shot down in court (equity). Also, companies are bought on a routine basis by buying all the shares - it's one of the common ways of doing a buyout. You bid x per share, which is normally a premium over what it currently is selling for on the market. Pay attention to business news and you'll see it a lot.<br> </div> Thu, 07 Jan 2010 04:17:23 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/368490/ https://lwn.net/Articles/368490/ giraffedata <blockquote> Isn't the company in question being bought, not bankrupt? That would mean that the buyer, ie google takes over contracts and responsibilities and thus can be sued. </blockquote> <p> The buyer doesn't normally do that. <p> A "company" isn't really something you can buy. Buying a company is an informal description of various kinds of transactions, but the most common one is that the buyer buys the assets of the company. E.g. Google buys the copyrights, patents, trademarks, real estate, machinery, customer lists, etc. You don't take over someone's liabilities just because you buy something (or lots of things) from him. <p> The seller is now a corporation whose only asset is cash. It dissolves itself and distributes that asset to its shareholders. <p> Another way is to buy all the stock of the corporation. If the buyer does that, the corporation still has all its liabilities, and by extension the buyer does. But these liabilities are why people don't like to buy businesses that way. And they're also a principle reason that the business was set up as a corporation in the first place. Sun, 03 Jan 2010 18:44:15 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/368481/ https://lwn.net/Articles/368481/ misiu_mp <div class="FormattedComment"> Isnt the company in question being bought, not bankrupt? That would mean that the buyer, ie google takes over contracts and responsibilities and thus can be sued.<br> </div> Sun, 03 Jan 2010 10:24:12 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/368308/ https://lwn.net/Articles/368308/ giraffedata <blockquote> <blockquote> There would also be no refunds to customers who had already paid for the "professional" service. </blockquote> I have a hard time imagining AppJet thought they were going to get away with that, at least for any amounts that had been pre-paid beyond the scheduled service end date. To do otherwise would invite a lawsuit, and worse. </blockquote> <p> You can get away with anything by dying. My understanding is that AppJet is a corporation that will cease to exist and there will be no one to sue. You take a big risk extending credit to a corporation, which you do when you pay for services in advance. There are corner cases where you can get officers, directors, and stockholders of a corporation, and people who continue the corporation's business, to pay its debts, but normally it's just the creditor's risk. Thu, 31 Dec 2009 01:15:13 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/368292/ https://lwn.net/Articles/368292/ utoddl <em>But it doesn't have a vi mode, which is pretty much a show stopper. :P</em> <p>Which is the show stopper, having a vi mode, or not having one? (I could go either way.) Wed, 30 Dec 2009 20:42:21 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/368222/ https://lwn.net/Articles/368222/ zooko <p>I would love to have something like EtherPad targetting <a href="http://allmydata.org">Tahoe-LAFS</a>! (I'm one of the Tahoe-LAFS developers.) If the EtherPad client-server protocol is a RESTful HTTP protocol then this is easier. (I retargeted <a href="http://www.tiddlywiki.com/">TiddlyWiki</a> to target Tahoe-LAFS by REST and it was <a href="http://allmydata.org/source/tiddly_on_tahoe/trunk/tahoe_tiddly/HTTPSavingPlugin.js"> pleasingly simple </a>.) I looked briefly at Bespin, but it kind of seemed like the client-server protocol was non-RESTful and would be a lot more work to adapt. <p>I see that this article says about EtherPad: "The actual code implements an EtherPad server running as a stand-alone HTTP server on port 9000." That's promising! I'm afraid I definitely don't have time to investigate this myself right now (I'm working on <a href="http://allmydata.org/trac/tahoe/milestone/1.6.0">Tahoe-LAFS v1.6</a>), but if anyone else wants to investigate, I would love to help as I can. Join <a href="http://allmydata.org/cgi- bin/mailman/listinfo/tahoe-dev">the tahoe-dev list</a> or #tahoe-lafs on irc.freenode.net. Tue, 29 Dec 2009 18:35:22 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/367943/ https://lwn.net/Articles/367943/ butlerm <div class="FormattedComment"> &lt;em&gt;There would also be no refunds to customers who had already paid for the <br> "professional" service.&lt;/em&gt;<br> <p> I have a hard time imagining AppJet thought they were going to get away with <br> that, at least for any amounts that had been pre-paid beyond the scheduled <br> service end date. To do otherwise would invite a lawsuit, and worse.<br> </div> Thu, 24 Dec 2009 00:43:47 +0000 anyterm? Ajaxterm? https://lwn.net/Articles/367934/ https://lwn.net/Articles/367934/ dmarti You could just create a regular user account, run regular Emacs, and show the user an anyterm or Ajaxterm terminal. Would that be cheating? Wed, 23 Dec 2009 21:48:41 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/367927/ https://lwn.net/Articles/367927/ gidoca <div class="FormattedComment"> But it doesn't have a vi mode, which is pretty much a show stopper. :P<br> </div> Wed, 23 Dec 2009 20:23:35 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/367923/ https://lwn.net/Articles/367923/ lambda There already is <a href="http://www.ymacs.org/">Ymacs</a>. It doesn't have an elisp interpreter, but it does implement a lot of the basic editing functionality of Emacs. Wed, 23 Dec 2009 19:51:51 +0000 EtherPad source code is free, now what? https://lwn.net/Articles/367922/ https://lwn.net/Articles/367922/ ajb <p>Well, I guess one answer would be to try and implement such services on top of a distributed backend such as <a href=http://freenetproject.org/>Freenet</a> or <a href=http://allmydata.org/trac/tahoe>Tahoe</a>. </p> <p> On a less serious note, I wonder how long it will be before someone ports emacs to the browser. Write an elisp bytecode interpreter and you've got about two thirds of it done :-) </p> Wed, 23 Dec 2009 19:11:25 +0000