Talk:Portable Document Format

From Wikipedia, the free encyclopedia

This is the talk page for discussing improvements to the Portable Document Format article. This is not a forum for general discussion about the article's subject.
Please sign and date your posts by typing four tildes (~~~~). Put new text under old text. Click here to start a new topic. New to Wikipedia? Welcome! Ask questions, get answers.	Be polite Assume good faith No personal attacks Be welcoming	Article policies No original research Neutral point of view Verifiability

This article is within the scope of the Graphic design WikiProject, a collaborative effort to improve Wikipedia's coverage of Graphic design. If you would like to participate, you can visit the project page, where you can join the project and see a list of open tasks.

Archives /Archive 1: Oct 2002 - Sep 2005 /Archive 2: Sep 2005 - Dec 2006

Contents 1 "Further criticism" 2 Reverse engineering a pdf 3 Backwards Compatibility 4 Openness and old Microsoft rumors. 5 Comparison of layout engines (XHTML) see also link

[edit] "Further criticism"

I was tempted to remove the following paragraph, but there may be something in there, which I do not understand:

"Further criticism is also centered around the restrictive, proprietary nature of the format, which does not allow free, direct encoding and modification of documents in PDF."

What does "free" mean here? Surely with a text editor, you can go in and modify any PDF file, as much as you want. You cannot usually add a paragraph to a text, and I think it is valid to mention that restriction, but that has nothing to do with the fact that the format is proprietary.

What does "direct encoding" mean? Are we talking about encryption or character set? If it is encryption, it seems the "criticism" simply is that the encryption algorithm is not free. But then, no one "criticises" Word for not being free, do they? I do not see any problem with the character set in later PDF formats, even though it could be messy in earlier version. Mlewan 08:19, 9 January 2007 (UTC)

This seems, to be honest, to be nonsense. If the comment is that PDF files are hard to directly edit and create (without using software), because of the complexity of the design of PDF, this is true (relative to simpler formats), but is unconnected to the proprietary origins. If this is stating that it is difficult because it is proprietary, this makes no sense: the PDF specification is readily available. If it is saying that because it is proprietary, free/open source software is not allowed, this is just wrong. So I have removed it. Of course, if I've completely misunderstood a valid and verifiable criticism, it only needs to be added back with a suitable source. Notinasnaid 09:30, 9 January 2007 (UTC)

Is it really relevant whether the criticism is valid? All that would be needed is a reliable source showing that this is a common criticism. Those who think it's a bad criticism (and I include myself) could cite sources refuting it. I agree with the deletion but I think it's an unnecessary distraction to argue about whether the criticism is correct; WP:NOR is a sufficient reason to delete it. Mrhsj 16:51, 9 January 2007 (UTC)

Quite right. Verifiable is enough; I should not have said "valid". However, it does need to be comprehensible. Notinasnaid 17:47, 9 January 2007 (UTC)

There are two potentially valid points that might be made. It is possible that the basic internal file format of pdf is such that it is difficult to modify/edit the contents. And it is very likely that part of the current wave of security features of all sorts is to intentionally make it extremely hard to edit/modify the contents of pdf files that use such security features.--69.87.203.22 16:42, 12 February 2007 (UTC)

[edit] Reverse engineering a pdf

Something about how to reverse engineer a pdf back to the point where it was saved as a pdf would be useful on this page. People are always asking me if they can do this; as i have only a hazy idea that there are tools out there that can do it I usually sidestep the question. Daveches 10:33, 24 January 2007 (UTC)

That isn't relevant to the article, which just describes the file format. There are plenty of tools which can work directly with PDF files. Chris Cunningham 11:40, 24 January 2007 (UTC)

The buzzword you want is "extraction": you're just getting out content, not reverse engineering which means something quite different. Notinasnaid 11:52, 24 January 2007 (UTC)

The buzzword is not necessarily "extraction": it's often about fomatting as well as content. eg if I have Microsoft Word on my machine and the PDF was saved from Word, can I get back to the Word document with all its formatting (eg to make a change then resave as PDF?)Daveches 21:37, 26 February 2007 (UTC)

There are commercial tools that will attempt to convert a PDF to Word format. They generally do not do a perfect job. The fact that the PDF format typically does not preserve all the information required to represent its source word processing document is germane to the article and it would be reasonable to include a statement about this if someone want to take the time to write a good well-sourced one. But a discussion of the capabilities of particular tools would be off-topic - this isn't a tech support forum. Mrhsj 04:56, 27 February 2007 (UTC)

[edit] Backwards Compatibility

The sentence about backwards compatibility should probably be deleted or changed. 1) No Backwards_Compatibility means newer readers can't read (some) older files, not the other way around - what the article currently claims. 2) Backwards Compatibility is almost always a function of the particular reader software used, not the format itself. Because this article is about the format, the statement is very confusing. I've also found no reference to backward compatibility issues online, and I'm inclined to think that the original author of that sentence was just frustrated because old reader software doesn't open new PDF documents; but that isn't really a backwards compatibility issue. Dana Powers 16:18, 1 March 2007 (UTC)

The sentence originally read "forwards-compatible." I wrote that, and that's what I meant. Someone changed it to backwards-compatible, which is incorrect. To me, saying "the format is backwards compatible" means that a reader written to a new specification will be able to open files written to an older specification. To say "the format is forwards compatible" means that a reader written to an older specification will be able to read a file written to a newer specification. The PDF format is backwards compatible; it is not forwards compatible. Mrhsj 17:14, 1 March 2007 (UTC)

Ahhh - which makes way more sense. It should definitely be changed back to forwards compatible then. Right now its just a logically confusing sentence. A cite would also be nice, as would clarifying that this is a property of the specification itself, not just a single application. Dana Powers 18:53, 1 March 2007 (UTC)

I do not get the point. Is any format ever forwards-compatible? Not even common text documents of today can always be opened in old versions of vi, because text files now often are UTF16. Check out .doc, .xls, .123, .rtf, .psd, .tif... Is any of them or any other format forwards-compatible? Mlewan 18:57, 1 March 2007 (UTC)

It is extremely difficult to achieve forward-compatibility, to be sure, which explains why most of the document formats you mention are not forwards-compatible. But it isn't impossible. Dana Powers 19:51, 1 March 2007 (UTC)

I've investigated a little further and found the BX and EX operators (this in the 1.3 spec; I haven't looked at others). They allow a PDF writer to specify that certain commands in a particular document should be ignored when read if they are unrecognized. This would allow creation of forwards-compatible PDFs, but it doesn't require it. Dana Powers 19:51, 1 March 2007 (UTC)

Yes. There are features in PDF that were intended to facilitate forwards-compatibility. That's why I rewrote the statement that said the format "wasn't designed" for forwards compatibility - it kind of was, it just didn't fully achieve the goal. I don't have a copy of the 1.0 spec so I can't do a proper sourced statement, but I recall that it promised that if a non-compatible change were ever introduced to PDF (one that would break an old reader completely) then the PDF version number would be bumped to 2.0. But such changes were in fact made (e.g. with the introduction of Flate in 1.2 and Object Streams in 1.5) without the major version being bumped. Mrhsj 05:13, 2 March 2007 (UTC)

[edit] Openness and old Microsoft rumors.

I deleted this material: "(There is now some contention whether PDF will truly be an open standard in light of Adobe Systems threatening to sue Microsoft Corporation for including the ability to write PDF's in some of their products, see external links.)"

Here's the history: in June 2006 Microsoft decided to change its PDF export tool from a bundled add-in to an add-in users would have to go their website to download. They claimed this had something to do with a threat of a lawsuit from Adobe, but this was never confirmed and no suit was filed. Then in January 2007 Adobe announced it was going to submit PDF to a standards body.

I don't see how those facts support the claim that "there is 'now' some contention...". That statement would require citations showing that there has been controversy *since* the Jan 2007 announcement, and that that controversy had something to do with this old rumor. Mrhsj 04:10, 15 March 2007 (UTC)

Agreed. The reported existence (or threat) of this antitrust suit has been widely used to support all manner of tenuous conclusions, in this article and elsewhere. Notinasnaid 08:43, 15 March 2007 (UTC)

It may be worth noting a particular error: some people assume that because Adobe is reported to be considering a suit because Microsoft may add PDF creation to Office, that therefore it must be because PDF is a secret, or requires a payment or license to develop for, and Microsoft don't have Adobe's permission for this. To be clear, no permission is needed, PDF is not a secret, and these are not the grounds for the reported threat of a law suit. Notinasnaid 08:56, 15 March 2007 (UTC)

I suggest the you STOP removing any discussion of the openness of PDF until you can point to ONE article from ANY source where a representative of Adobe states that they have no intent to sue Microsoft (or any one else) for inclusion of PDF code in Office 2007 (or any other product). This smacks to me of someone who has a strong vested interest in Adobe and is not interested in unbiased reporting of the facts. It should also be noted that handing a specification over to a standards body does not in itself make the standard open in any real sense. Please review the situation with the JPEG standard which which was built from the ground up as an open standard, but was the subject of extensive litigation years later. Forgent Networks sued over 30 corporations for patent infringement for creating applications that could read and write in the JPEG "standard".

So, come forward with VERIFIABLE proof of your claims that PDF is an open standard with no threat of litigation, or stop removing discussion of this issue. It is critical information to anyone that is considering adding PDF support to an application they may write. Your whitewashing of the issue does not help. - Sincerely, a non-Adobe employee. —The preceding unsigned comment was added by 63.100.100.5 (talk • contribs) 22:25, 20 March 2007 (UTC)

I believe you are mistaken. The burden is not on editors to show that that Adobe have denied an intention to bring a suit against Microsoft. The burden falls on an editor who wants to add information that it is verifiable. The suit has been reported, but drawing conclusions about the openness of the standard from the suit seems to be entirely different, and this requires a reliable source. Reports I have read sugest the suit relates to anti-trust issues, not intellectual property or standards. You also seem to be confusing open standards with patent issues; these are entirely different things. Some open standards contain acknowledged patents and some are alleged later. However, there is disagreement on the definition of "Open standard", please see the article for details. If you are interested in Adobe's statement on the intellectual property issues connected with PDF, please see section 1.5 of the PDF Reference. Notinasnaid 22:52, 20 March 2007 (UTC)

I still disagree, my points are very well documented. There are at least 10 articles from reliable sources that discuss the pending lawsuit, but NONE that state there is no threat of lawsuit. How can you continue to state that the "burden falls on an editor who wants to add information that is verifiable". My information is completely verifiable, yours is not. Also you state that "suit relates to anti-trust issues, not intellectual property or standards". Well, it doesn't matter what the suit relates to directly. It directly influences the format and is critical information that someone who comes to Wikipedia might need to know (e.g. if they are deciding if they can add PDF export to their product). If you still disagree I suggest an independent third party review what is verifiable and decide if this information should be shown in the wiki entry. So far I see little to convince me that you aren't just an employee of Adobe who has a strong vested interest in showing that company in a positive light. I don't work for Adobe, Microsoft, or any other company that has a vested interest in the PDF/Markup business. However I strongly suspect that you do. —The preceding unsigned comment was added by 63.100.100.5 (talk • contribs) 23:41, 20 March 2007 (UTC)

Please try to rein in the personal attacks and let's focus on the article content. If this article reports the lawsuit it can only report what the reliable sources say. If the source draws conclusions, it can report those conclusions (as quotes, if they are not universally agreed). But if editors draw those conclusions, this is original research and clearly not allowed. My view is that what people planning to add PDF export need to know is in the PDF Reference, as defined above, do you have a reliable source that contradicts this Adobe source? Now, I'm going to wait and see what others say. Notinasnaid 00:17, 21 March 2007 (UTC)

I'm not making a personal attack, but I am questioning whether you are impartial. I think it is quite well established that impartiality is critical when providing public information. I believe you should state unequivocally whether you are financially compensated by Adobe so other readers can fairly judge your motivations. This becomes a critical issue over time since someone who is paid can keep arguing forever, whereas someone who wishes to add content because they believe it to be accurate and in the best public interest to distribute will eventually give up. Whatever the case, I hope a disinterested third party will make a judgement so I can move on. If they say I'm wrong and the information should not be present, I'll drop it immediately, but if they say the content should be present I think it should be added and you should be prevented from removing it. —The preceding unsigned comment was added by 63.100.100.5 (talk • contribs) 03:36, 21 March 2007 (UTC)

Well it's been a couple days and no third party has given an opinion. Unless someone offers an opinion soon, I'm going to add content discussing the lawsuit back in since it appears to be a simple difference of opinion and there is no consensus. Unless someone else rules, I expect that the content will be left in the article. Twikir 08:35, 22 March 2007 (UTC)

The rationale for removing it is still valid. This isn't a case of consensus, it's a case of policy, which says that random unsourced or poorly-sourced contentious material has no place on Wikipedia. (out of interest, are you the same user as the anon above?) Chris Cunningham 09:18, 22 March 2007 (UTC)

As I see it, the contention is whether the threat of suit by Adobe over Microsoft's attempt to add PDF creation functionality calls into question the concept of PDF as an 'open standard'. Do we have a reference that actually makes the above supposition? If so, put it in the article with the citation of the source and let's be done. If not, then the supposition is original research and it shouldn't go in. Whether Adobe actually has any intent to actual start litigation seems irrelevant to the issue at hand. Blade 15:43, 22 March 2007 (UTC)

I'm not sure if the question was directed at me, but since I'm the one who removed the material in question, I'm happy to state that I do NOT have a conflict of interest on this topic as defined by Wikipedia. Mrhsj 21:06, 22 March 2007 (UTC)

[edit] Comparison of layout engines (XHTML) see also link

Is the "Comparison of layout engines (XHTML)" link under see also really relevant for this article, or am I missing something? --ScottyWZ 04:32, 17 March 2007 (UTC)