E Text Editor: Releasing the Source | Hacker News Reader

E Text Editor: Releasing the Source

e-texteditor.com

51 pointsbdfh4222 comments

I have had a lurid history with E:

Basically this was the only program I could find which was similar in workflow to TextMate and provided in-program FTP/SVN support (with a few hacks).

However, the program was so poorly designed and so buggy that it made working extremely tedious and dysfunctional. It made me start thinking about how somebody could release a program so ridiculously buggy and what were the underlying reasons. While the programmer is the first obvious target, I realized that the Windows framework was probably to blame. It's not the framework itself, but the way it makes you code. It was hard for me to justify continuing to use such a flawed system, as it scared me to think that potentially all these ill effects may be rubbing off on me. As a result, I switched to Mac the next day.

Therefore, E texteditor made me lose enough faith in Windows and Windows programs, which I had been using since 3.1 (even using 1.0 at one point in 1995), that I switched my entire operating system. I haven't looked back since. It may have been coming anyways, but it was a good decision.

While the programmer is the first obvious target, I realized that the Windows framework was probably to blame.

It's not the Windows framework, it's wxWidgets, an open-source cross-platform framework.

To be fair, though, it's been designed to be like MFC.

Sorry to hear you had such a bad experience. Most users of E suffer from a few bugs because the developer likes to get release early and release often, so bugs often creep in, but they have mostly been fixed very quickly. I'm not sure why you should have had such problems. I've used E for a couple of years now and have only been troubled a couple of times; generally it's been very stable.

That's a very curious license.

So you have all the source, but if you build something based on it, only licensed owners of E can use it?

edit: I take it back, you do not appear to get the source to ecore.lib, which appears to contain licensing code and perhaps other things.

Yet, there's nothing in the license keeping you from downloading it, removing the license code yourself, and running it registration-free. (Your personal sense of morality may differ, I suppose.)

For curiosity's sake, what would be the legality of a patch to remove said licensing code from the code?

I'd worry it might run afoul of the DMCA since it would be a tool for removing copyright protection

Normally I'd interpret it like you have, but the license specifically says that only redistributions must keep the licensing in tact. I don't know if that was an intentional thing on their part or just a mis-wording that they didn't mean, but it definitely says that you have to keep the licensing in tact if you redistribute it.

I'm guessing that it was intentional. There are some people willing to go through the effort to remove the licensing from the source and compile. Who cares! The real issue (in terms of losing profits) is if one person is able to remove the licensing and redistribute that change.

A patch to remove a registration check is arguably a tool for copyright infringement no matter what the license says (the author is still the copyright holder after all). But I'm certainly not a lawyer.

Probably (Not a lawyer, etc) legal. There's versions of XChat for Windows that are built from the source and have no differences besides the licensing code stripped out of it.

XChat is different. XChat is free software (GPL), and the "official" windows code is a completely separate add-on with code that has never been released. The freely-available windows builds use different code, and as a result there is no licensing controversy whatsoever. It's a GPL'd derived work of a GPL'd project.

In fact, the author of XChat is likely violating the GPL by distributing the Windows version without full source code. Since he's accepted third-party patches under the GPL, he has to distribute all derived works under the GPL (or remove all the code that isn't his). This means that currently his closed-source Windows version is a GPL violation.

I know that if I had code in XChat, I would be upset by this.

I was considering doing this same type thing with one of my mac applications. My thought was, if they download it, remove the registration, and get it going without a license, at least they have the source on their computer and familiarity with it. The possible will make a good contributor. I was also thinking about creating a "developer license".

DJB says patches are always legal:

http://cr.yp.to/softwarelaw.html

Here's the explanation of why the license is the way it is, and why the code is only partially open:

http://e-texteditor.com/blog/2009/opencompany

I love E. In a way, with the split screen editing, it's even better than Textmate. For windows users, E is one of the best editors out there. Releasing E's source is an awesome move!

Their "Open Company" concept is rather interesting too. Almost a sort of e-Kibbutz.

Actual link to source: http://github.com/etexteditor/e/tree/master

Edit: I initially misidentified the project-specialized clause as the key new clause; my text below has been expanded and corrected. I thank almost and artost for their corrections below.

This project's eccentric worries don't justify the license proliferation.

The new elastic "do what I mean" clause -- "Any redistribution, in whole or in part, must retain full licensing functionality, without any attempt to change, obscure or in other ways circumvent its intent" -- just adds a new cognitive barrier to the easy participation and mixing with other projects that's usually the whole point of open source.

I also wonder if the new restriction is legally meaningful in any realistic situation. There was no right to arbitrarily relicense BSD code to circumvent its conditions before the addition.

Given that the essential conditions of the BSD license are attribution and non-endorsement, how does this new clause add anything but confusion?

Even in the absence of the updated wording, any marketing falsely implying a endorsement from the main project would be actionable under other principles. Even with the updated wording, is there cause of action against someone who, in their promotional materials, makes a factual statement about the origin of the code?

If so strictly enforced, the BSD license itself is almost self-contradictory: you must include a "Copyright (c) 2009, Alexander Stigsen, e-texteditor.com" in distributions and documentation... but you must not use his name or product name to "promote" your derived product. Can you achieve both of those when the distribution and documentation are themselves the primary "promotional" materials of a free work? Adding the new 'anticircumvention' wording only makes this worse; I wonder if a lawyer was even consulted in the redrafting.

And, it's not like the name "e-texteditor" or the names of the current or future contributors are marketing gold, anyway.

That's not the new clause, that's from the BSD license.

The new clause is the last one in the list, the one about not removing the license enforcement code.

You're right; <strike>but they've even reworded that clause.</strike> I'm correcting my grandparent comment to reflect reality, with a note that it's been changed.

In what way have they reworded the clause? As far as I can see it is exactly the same as in http://www.opensource.org/licenses/bsd-license.php

You're right too. I mistakenly referred to a non-parameterized generic version of a "BSD style" license to make that statement, as here:

http://www.debian.org/misc/bsd.license

My initial writing was also confusing the 'simplified BSD license' used by FreeBSD (without the "no-endorsement" clause) with the full classic BSD.

And this is another reason to hate license proliferation. Remembering all the "BSD, except X" variations is an error-prone pain. Let's be coding and sharing, not lawyering.

I remember trying this editor out when I was looking for something good on Windows before I switched to using Vim. It was well thought-out; very TextMate-like.

How does this not completely kill their business? Are they really expecting everyone to pay? Or is there only partial openness to the source?

edit: Just saw eli's comment that the core is in fact closed.

Reply on news.ycombinator.com