Older blog entries for cinamod (starting at number 53)

Hi rillian,

I'm also a big fan of Free interchange mechanisms, be they file formats, protocols, languages, etc. I don't think that PDF is evil. I just think that it's less warm-and-fuzzy than a lot of people would like to believe, and that it's not substantially warmer or fuzzier than the Microsoft formats, legally speaking.

You're correct that they don't have an open development process. As far as I know, there isn't even a driving/coordinating committee like we see with the W3C and OASIS groups. PDF documentation is good and freely available. But so is documentation on the Word 2, 5, 7, and 97 formats. Both specs have their share of flaws and inaccuracies. Both specs can be gotten readily enough online.

It's truly refreshing that Adobe has a blanket patent grant for conformant apps. This is nice, but it's not Free or Open, since I'm not free to make a derivitive work that is non-conformant and still use those patents and other copyrighted bits. Further, Adobe explictly forbids use of patent 5,860,074 in a viewer (Method and apparatus for displaying an electronic document with text over object) which seems quite core to writing a compliant PDF viewer. From my cursory reading of the abstract, they're forbidding viewers from rendering Linearized/"Web Optimized" PDFs, at least intelligently. Ouch.

The PDF spec does include other sub-components that are or were patented by !Adobe (such as LZW, JPEG, JBIG2, JPEG 2000, etc.). Rillian is right - getting sued over these things is probably a low-risk endeavour. But the risks and patents remain, thus being a stumbling block toward "Freeness". I also contend that a conformant viewer would have to support those patented bits in order to recieve Adobe's blanket patent-grant. And the PDF spec is a daunting beast. Producing a conformant viewer is nigh-impossible. I've tried more than once.

It's nice that PDF is easier to parse than PS, but that's not quite germane to our argument. Considering that PS is a Turing-complete programming language, it's not exactly hard imagining an easier-to-interpret format. And it's worth noting that the majority of PDFs I encounter are corrupt in one way or another.

So, even if things are "nice" and "low-risk", it doesn't make them Free or even Open. Any way one looks at it, PDF the spec and PDF the format is a patent-encumbered, copyrighted entity that you're not allowed to distribute or make derivitive works thereof. Adobe has both said and shown that it's willing to enforce those protections when it suits them to do so.

I'm not saying that we shouldn't use (a subset of) PDF and make it our own. I'm pragmatic. To me, it looks like the best option out there today. I'm just saying that people like RMS shouldn't be hypocritical when they advocate PDF over another format. On "Freeness" and "Openness" issues alone, we might be better off taking a subset of the equally well-documented, widely-available, patent-unencumbered DOC format and make it our own. But I honestly don't want to see that happen.

21 Jun 2005 (updated 21 Jun 2005 at 15:20 UTC) »

I'm all in favor of the ability to be selectively anonymous, both online and in real life. But I do take a bit of offense when people write out otherwise well-intentioned, thoughtful emails that deserve equally thoughtful responses, but I can't really reply due to their anonymity. There's no need to hide behind the curtain - I'm not about to judge you or think less of you because you were committed enough to take a public stand. If anything, I'd admire you moreso.

In this case, someone used Cypherpunks (Anonymous via the Cypherpunks Tonga Remailer <nobody@cypherpunks.to>) to respond to my last blog entry. Personally, I'd prefer that discussions that start out public stay public. So I'm going to make portions of this mystery man's statements public in order to respond to them. I hope that my doing so doesn't upset this anonymous OSS denizen.

Like many OSS people, this well-intentioned poster confuses "Open" with "Free". The poster also confuses "open" with "available for download". Since RMS and the FSF concern themselves with Free (and publicly and adamantly disavow/distance themselves of "Open"), my previous and current arguments focus on asking how the Free software founder and Foundation can advocate using a non-Free format over another (perhaps equally) non-Free format.

I also want to say that I have some intimate knowledge of both the PDF and DOC standards and their licenses, having worked at Appligent on wvWare, AbiWord, GnomePrint, etc., and having had some IP law training. With that said (and having just given myself enough rope to hang myself with), let's get to the email.

[ANON] Yes, Adobe owns PDF but 1) the specs are open - and remain open.

The specs are available and remain available, and I never claimed otherwise. I remarked that the spec is not Free. I'll also show that the specs are not Open. And, given the copyright and restrictions Adobe has placed on the format itself (as we'll see in later paragraphs), it is doubtful that a PDF reader or writer could truly be Free or Open as well.

[ANON] 2) Adobe freely licenses the format, in a manner that is consistent with all open source licenses.

I don't think that the format's or spec's license is consistent with any OSS licenses (re-distribution of the spec is forbidden), let alone all of them. That is an awfully big brush to paint with. What's certain is that Adobe publishes the format, but the terms under which it does so (and the conditions under which one might make software to implement said format) is not consistent with the GNU licenses, which is the inconsistency I allege. In my previous blog, I made no reference to the spec being Open, merely it not being Free. But it is neither. I quote from the PDF 1.5 reference:

NOTICE: All information contained herein is the property of Adobe Systems Incorporated. No part of this publication (whether in hardcopy or electronic form) may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of Adobe Systems Incorporated. Please note that the content in this guide is protected under copyright law even if it is not distributed with software that includes an end user license agreement.


This limited right to use the copyrighted list of data structures and operators does not include the right to copy this book, other copyrighted material from Adobe, or the software in any of Adobe's products that use the Portable Document Format, in whole or in part, nor does it include the right to use any Adobe patents, except as may be permitted by an official Adobe Patent Clarification Notice (see the Bibliography).

That's in reference to the documentation itself, which is neither GFDL, CC, Open, or public domain. It's not even close. The SPEC itself is not Free - it is the wholly owned, proprietary property of Adobe. The spec is not Open, as the terms and conditions of its license/copyright still enjoin you from creating derivitive works of that spec or even re-distributing the spec, unaltered. Later on page 7 of the 1.5 reference:

The general idea of using an interchange format for electronic documents is in the public domain. Anyone is free to devise a set of unique data structures and operators that define an interchange format for electronic documents. However, Adobe Systems Incorporated owns the copyright for the particular data structures and operators and the written specification constituting the interchange format called the Portable Document Format. Thus, these elements of the Portable Document Format may not be copied without Adobe's permission. Adobe will enforce its copyright.


Authors of software that accepts input in the form of the Portable Document Format must make reasonable efforts to ensure that the software they create respects the access permissions and permissions controls


Anyone who uses the copyrighted list of data structures and operators, as stated above, must include an appropriate copyright notice.

After that, Adobe gives some set of exceptions, meaning that one could create an implementation that reads or writes the format, within some (large-ish) set of constraints.

But notice how you must also enforce Adobe's (lousy, hole-filled) DRM standard if you wish to do so, as well as include a BSD-ish advertizing clause when you do so. Note that your right to make software around this "open" spec is a "limited right" that Adobe may rescind if they so choose. Note that the format may include patented bits which Adobe explicitly says you can't use.

Adobe owns and will police this particular, unique implementation of an interchange format. These elements of the Portable Document Format may not be copied without Adobe's permission. Adobe will enforce its copyright.

[ANON] Compare to traditional DOC: 1) no specs (pulled, as you state), and 2) no license.

These specs exist, and were published by Microsoft. The binary MSFT Office formats have not changed substantially since 1997, when the last specs were published. As far as I know, the specs are under the same license as all content on MSDN. There are Free and Open implementations of the spec (wvware, antiword, catdoc, wv2, OpenOffice.org, etc.).

[ANON] Compare to the new Word XML: 1) specs are provided, but 2) the license invokes patents and many believe that clauses in the license are designed to be incompatible with some open source licenses including GPL.

Adobe's license also invokes patents and clauses which are incompatible with the GPL. The license and spec have many clauses that are designed to prevent Openness and Freeness. When you take into account the evidence I've provided, one can only conclude that PDF is not substantially more Free than DOC, despite any warm feelings one might have toward Adobe or hostile ones toward Microsoft. I'd argue that PDF isn't particularly even more Open than DOC. That's not my concern, but I think that I've proven that anyway. My concern is merely to show the inconsistency of the RMS anti-DOC stance, when one takes into account his proposed PDF alternative.

Also, please note that Adobe has invoked the DMCA to stop people from reading their own, purchased eBooks, which are essentially encrypted PDFs. This company is not a big, cute, cuddly teddy bear, and if I were RMS or the FSF, I wouldn't want to advocate in favor of using one of their standards simply because of this point.

So in conclusion, PDF is non-Free. DOC is non-Free. RMS says he "can't" read DOC, but "can" read PDF. I contend that this is inconsistent behavior. Either RMS should stick to his guns and discourage use of PDF as well (or better still, convince Adobe to free the PDF spec), or he should bite the bullet and use wvware, antiword, catdoc, or some other GPL'd program/library to read MSFT Office documents. Anything short of that is tantamount to hypocricy.

I'm not arguing in favor of boycotting Adobe, using Microsoft products, vice-versa, or anything in-between. Mine is to only point out the inconsistency in RMS' "use PDF and not DOC because DOC is eeeviiil" paper. If you're going to argue for Freedom, stick to your guns. If you're going to argue for Openness, stick with that. I don't think that Adobe's spec or format meets either criteria. But having written things that read and write both PDF and MSFT Office formats, you can guess how much I do or don't care about the formats and their specs so much as the ability to inter-operate with the things that generate and consume these formats. Reform comes in excruciatingly small increments.

Anyway, thanks for the thoughtful email, and I'm quite happy that you enjoy using AbiWord! If you're a FOSS hacker or otherwise a cool cat, thanks for keeping-on keeping-on.

Hub, Steven Garrity perhaps draws the correct conclusions, but he draws them from the wrong facts.

  1. The older (binary, non-XML) MSFT Office formats are not patented, to the best of my knowledge.
  2. The newer (XML) MSFT Office formats are relatively open (to the eyes, at least) and non-binary, but they have potential patent problems. Any such patent dispute would likely be found null-and-void in light of prior art (ahem, AbiWord, Gnumeric, KOffice, OOo, etc.), provided that you had sufficient legal defense in court. In fact, lawyers in New Zealand are using AbiWord's DTD and XSD to dispute MSFT's patent application there.
  3. The various MSFT Office formats have (largely) been reverse-engineered. Garrity has to admit as much when he talks about "workarounds", but he does not. This omission makes his previous section ring hollow.
  4. You do not need MSFT Office to read and write documents compatible with MSFT Office. They do not own the ability to access the works we've created. See my prior point.
  5. RMS and his followers are wrong to rant against MSFT Office documents while at the same time arguing in favor of using Adobe's almost equally non-open PDF standard. Those who quote or reference this particular work make me sad. Yes, there are PDF viewers for GNU/Linux. That doesn't make the standard Free or its various required subcomponents Free (ahem - LZW, JPEG, colorspace conversion, ...). The standard is still very much controlled by Adobe, who were nice enough to publish a few books on the topic. But then again, MSFT has also published the Word97 specification, even if they did later retract it. If RMS and his followers want to argue that MSFT and proprietary software/standards are immoral, great. I'll not only defend his argument, but I might even carry his flag. But it's insane to complain about how one "can't" read the documents. wvWare; antiword; catdoc. All GPL'd software. All work well. If you're going to be consistent with your PDF stance, pick one. If not, remove the PDF blurbs from your article. But you're not going to win over any converts with manifestos like that - you're only looking like a jerk with the folks that are trying to communicate with you.

I am now kitten enabled!. The next thing to do is to Gimp-up my photos and submit the cutest one to KITTENWAR. May the cutest kitten win!

From irc.gnome.org#librsvg:

<scarlet> so basically he's an arsehole yeah?
<scarlet> I'm going to be subtly rude to him to see what happens
<rahga> Hrm, reminds me of a joke.... If scarlet decided to be subtly rude, how would you know? ;)
<scarlet> Probably you'd work it out when you walk in on me sodomising your mother or filling your house with asbestos particles, but if you didn't notice that, I'm not sure how you'd know.
<dom> scarlet: i'm hungry. do you have any lead paint chips lying around?
<scarlet> you mean wall candy?

(kudos to Penny Arcade)

I have no fewer than 30 mosquito bites on my back. It's times like these that I wish that I were one of those disembodied heads in Futurama.

6 Jun 2005 (updated 6 Jun 2005 at 20:38 UTC) »

First, Apple announces that it's going to transition to using Intel-based chips by 2007.

Then, Debian "Sarge" is released.

/me checks the calendar - not April 1st.

Ok, we now have confirmation of 2 of the 3 signs of the apocalypse. The NOAA reports that hell's present temperature is approximately 0 Kelvin. The FAA is warning you to please beware of flying horsemen. That will be all.

I'd like to thank Steven Garrity for redoing P.G.O.. Most of my criticisms have been addressed. It:

  • Grows and shrinks nicely, within reason
  • Looks ok in MSIE
  • Loads and renders pretty quickly
  • Doesn't resize images unevenly if an img is inside it. The image just overruns its column's border, which I think is preferable.

I also love the look. I'd like to retract my previous "column is too narrow" statement. I think that the width is pretty good, especially since the column can resize intelligently. Thanks Steven.

Havoc: Please re-read, as I'm making 4 distinct arguments. They are:

  1. I don't like the aesthetic, independent of the real estate issue. I still contend that it looks cool for 10 seconds, and then the novelty wears off quickly.
  2. It unnecessarily wastes loads of real estate without conveying useful information or increasing usability. I contend that the majority of wasted space exists not for usability issues, but "hey, floating heads with speech bubbles look cool" issues.
  3. The page is unecessarily heavyweight relative to what information it conveys.
  4. It doesn't work in any browser other than a recent version of Firefox.

I've never argued for text to go across the entire screen. But covering more than 50% of it might be nice. Instead, we're left with large vertical and horizontal dead zones that don't appeal to my sense of style and negatively impact how quickly I read the text. Western languages read left to right, top to bottom. Every carriage return means that the eye has to backtrack and scan down one line. There's probably a happy medium here. I just don't think that what we've got is it.

The most commonly used word processing usage is what Microsoft calls the "Normal" mode, which shows only the editable page area, without any page margins.

Finally, your analogy to newspapers is weak, as they don't include the author's head in its own column. Pick up any paper or online journal - the author's head is almost always top left, and less frequently, top right. They don't devote whole columns to the author's head.

Again, just my $0.02.

23 May 2005 (updated 23 May 2005 at 19:23 UTC) »

[Update: A few people asked me to clarify my objections to the new PGO site]

I appreciate the work being put into a new layout. I must say that as aesthetics go, I prefer the old layout to the new one. IMHO, the speech bubbles look cool for about 10 seconds, but the novelty wears off quickly. But there's no accounting for taste, or my lack thereof. After that, I start caring about things like:

  • Minimizing wasted space. I think that the columned text with large whitespace on either side leads to decreased readability.
  • The increase in time the page takes to download and render, since the new site uses lots of images to achieve the speech bubbles and drop shadows.
  • The page working well in any browser, not just recent versions of Mozilla/Firefox. The new site layout looks like smeg using Mozilla 1.6 and earlier, or Internet Explorer 6 and earlier. This doesn't mean that the page must render identically in every browser, just that it must be reasonably readable in any browser. I'm willing to provide more screenshots if you care to correct the problem. For what it's worth, it does seem to look ok in lynx :)

But, you can't please all the people all of the time. Please take my opinion with a grain of salt.

44 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!