I hope that Germany mandating ODF over OOXML will enhance the whole ecosystem.
As a programmer, finding decent ODF libraries is far from certain. Last year I had to output some spreadsheets from a Go program, but I could not find any maintained library for ODS, so I had to output XLSX files. Recently, I was luckier while programming in Rust.
LibreOffice needs to pull their heads out of the sand and recognize that both OOXML and ODF are the past.
Make it a polyglot ZIP à la SingleFileZ if you have to (or polyglot JSON[1] or just straight up XHTML with a big sidecar blob of embedded metadata), but nothing trying to take on Microsoft Office is "the future" if it's trying to get there with a strategy shackled to the notion of people downloading the appropriate format-compatible software for something as simple as being able to view (not even edit!) the document that has been sent to them.
It's great to have open standards and free software, but be compatible by default with the universal formats understood by the readers for the ubiquitous document infrastructure that everyone already has installed (WHATWG-/W3C-compatible hypertext readers, i.e. Web browsers), or be forever doomed to the same level of obscurity that OpenOffice/LibreOffice were fated 10–15 years ago due to their myopia.
> A forward-looking format is one that reduces future dependency, not one that reinforces it[…] A “backward-looking” format, by contrast, is one that ties the future to the commercial strategies of a single vendor. In this sense, OOXML Transitional is an archaeological artefact that preserves the past at the expense of the future.
This is delusional. As long as it's modeled on the same outdated paradigm of 80s- and 90s-style office suites (no matter which one), then any format whether it's an ISO standard or not is hopelessly attempting to preserve the past.
> Even today, web formats cannot render documents with the same fidelity as ODF. Especially not for spreadsheets.
That's a nonsense claim—literally a category error. You're conflating the limitations of what can be represented in a given file format with an implementation's (lack of) support for doing things encoded in that format.
LibreOffice doesn't render documents encoded in plain-text, browser-compatible formats with the same fidelity as ODF because LibreOffice has not implemented encoding documents in plain-text, browser-compatible formats that it renders with the same fidelity as ODF.
There is nothing in ODF that's inherently unrepresentable in JSON or pure XML.
If you try to print any non-trivial Web page with any of the existing Web browsers, all of them will print different things and none of them will print the same thing that they render on the computer display. For all of them the printed files will have various obvious defects, caused by incorrect size ratios between various elements or by an incorrect stacking of the elements, so that some of them obscure others that should have been visible.
For instance, Firefox and Chrome almost always print garbage, while Vivaldi is typically much better, but it also fails from time to time. It looks like all those who maintain Web browsers only test how pages are rendered on displays, but they never test how they are rendered by a print command.
With a file format intended for documents, the first property that I demand is that the document will be rendered perfectly in a deterministic way and it will look absolutely the same regardless on what medium it is rendered.
Nothing could be less true for anything that is used on the Web.
I strongly hate anyone who provides documentation in the form of Web pages, instead of using real documents, e.g. PDF files, ODF files or even Microsoft Office files, which can be used offline without problems. Nowadays, even the attempt to save Web pages in browsers is unreliable, due to the embedded scripts that may fail to work offline, making impossible the rendering of the saved pages.
> I am not sure what you claim about web formats.¶ If you try to print any non-trivial Web page with any of the existing Web browsers[…] For instance, Firefox and Chrome[…]
... why do you think this matters? Like, at all? We're talking about LibreOffice.
> Nothing could be less true for anything that is used on the Web.
This is, uh, crazy. In the first place, PDF is used on the Web, and it's both better suited for print-perfect sources and it's more portable. In the second place, believing that there's some magic inside ODF (or some other office suite's native format) that can't be represented in XML, JSON, or some other plain-text format is delusional.
If you claim that some text format used on the Web is suitable for storing a document, you should be able to point to some application that demonstrates the use of that text format in a manner that can be used for a document.
If none of the existing Web browsers can render a Web page in a predictable way in all circumstances, that clearly disqualifies all "Web formats". If the most complex programs like the Web browsers cannot always render correctly a Web page, then who can?
> There is nothing in ODF that's inherently unrepresentable in JSON or pure XML.
Since ODF is XML, that is a tautology.
I do not doubt that it is likely that ODF is more complicated than really necessary, and that something better could be designed.
However anyone who criticizes it should define an alternative better format and describe its advantages.
Pointing to some of the existing markup text formats is not this, because those lack a great number of features that are absolutely necessary for any document file format, so they are not comparable at all.
Saying that a markup text format can be suitable as a document format resembles the claim of the Americans that ASCII should be good enough for the speakers of non-English languages.
It would be great if you would stop and at least try to understand the argument you find yourself in and what's being claimed/described by the other side, especially in the comments you're choosing to respond to, instead of hallucinating some altogether completely different claims, like whether any two Web browsers consistently render HTML- and CSS-based Web pages in the wild the same—as if that's somehow pertinent.
> If none of the existing Web browsers can render a Web page in a predictable way in all circumstances, that clearly disqualifies all "Web formats"[†]. If the most complex programs like the Web browsers cannot always render correctly a Web page, then who can?
There is no brilliant observation here. We are not talking about how "the existing Web browsers can render a Web page in a predictable way in all circumstances". The answer to your question literally doesn't matter.
We're talking about the choices available to LibreOffice—choices about how to encode documents in a bytestream that can still be processed by apps that don't have explicit support for formats like ODF, and processed by those that do (apps like, you know, LibreOffice) that's (a) done in a way that's still consistent with the way that it (again: LibreOffice) treated the content the first time around, before it wrote it out to disk (b) consistent with the way that other apps that commit to the same standard of interoperability treat that content. Those are the _only_ tests that matter in this conversation. You can imagine all sorts of other tests and cross-examine against them, but no matter what you manage to come up with, they do not matter.
It's your insistence on not understanding this that's the problem.
The questions—again, the _only_ questions—to answer are:
1. Can LibreOffice encode its documents in such formats and in such a way that it doesn't compromise its ability to have the documents cleanly "round tripping" their way out (of LibreOffice) and back in (to LibreOffice) again—i.e., whether it can use one of these formats as its container (rather than the ZIP-for-XML-&c container that ODF uses). The answer to that question is an unequivocal "yes"—there is literally _nothing_ magical about ODF that can't be captured/encoded using some other scheme for representing data. It's equal parts baffling and maddening that this needs to be stated so, so, so, so, so many times in this discussion.
2. And whether LibreOffice—and any other desktop publishing app that aims to be interoperable—can render those documents "in a [reliably] predictable way in all circumstances". The answer is "yes—assuming it's able to do that now with ODF, and you aren't just putting your thumb on the scale in two dimensions [instead of just the one, as you've been doing throughout this exchange]".
Any and all other of these asinine tests and complaints about the deficient behavior of non-LibreOffice apps (like Firefox, or Chrome, or Windows 95's calc.exe…) are all half-thought-out red herrings that have next to no bearing on the topic at hand. Just mindnumbing attempts to move the goalposts.
† As I already pointed out in my previous comment, and which you inexplicably ignored, PDF is, in fact, quite good at this—and not even just that; it's so much better suited for what you're kvetching about.
> ODF is XML
Uh, no. ISO 26300[1], much like the Open Packaging Conventions[2] used for Microsoft's OOXML and XPS, "uses a package file to store the XML content of a document together with its associated binary data, and to optionally compress the XML content. This package is a standard Zip file". Leaving that aside and the monumental difference that that detail makes wrt the claim and its relevance to this situation:
There are no provisions within the spec requiring ODF producers make sure to spit out a ZIP bytestream that is carefully constructed so it can be processed by a pipeline that is expected to yield a meaningful result when it's treated as something other than ODF-flavored-XML-inside-ZIP (like application/pdf, image/png, image/svg+xml, or text/html, etc)—i.e. ODF packaging has none of the guarantees that SingleFileZ archives provide.
> However anyone who criticizes it should define an alternative better format and describe its advantages.
The media (web or desktop) is irrelevant: a file format must exists for backup and interoperability. I barely use office documents myself, but I work on software that produce and parse many spreadsheets every day.
An open standard is even more very relevant in public administrations where the process follows legal constraints and ISO standards. The Document Foundation's article reacts to an German institutional decision.
None of the documents that I write for myself and much less any of the documents that I write for others could be written as .md.
The reason is that I always use only good typefaces and I must use various typeface features provided by the OTF format, to which something like .md does not provide access.
If .md would be extended with the missing capabilities, then it is likely that it would become not much simpler than ODF.
Even in the simplest documents that I write for myself, I must use multiple document styles, as provided by a format like ODF, because I typically mix multiple languages, including multiple alphabets.
.md is OK for a ReadMe file written in English, but it is not good enough for what I would call a real text document.
Just last week I needed to output a formatted document, and my choices were constrained to: HTML+CSS, Markdown, docx (OOXML), or odt (ODF).
The formatting had to be manipulable in code; the first two options were unsuitable for my use case. That left OOXML and ODF.
Both docx and odt are zip files, with XML content. Any guess which one turned out to be the better choice? I was left wondering why docx even exists in the first place.
Anyone using OOXML can rightly be regarded as stupid. There are far superior formats, and we should insist on their usage.
At some point mandating a universal document format for ALL government use, becomes the only reasonable way forwards, along side tax software, utility billing, employee payrole,recipts, and banking that is also standerdised, and free.
Criminalising any further attempt to "patent" , obfusicate, privatise, or close source a public comunication medium as a shot accross the bow is the first step, with an expliset declaration that total dismantaling of companys attempting to insert themselves and comodify basic rights and privlages is going to happen.
Maps and navigation are also on the list.
Historical precidents are impossible to dismiss.
I hope that Germany mandating ODF over OOXML will enhance the whole ecosystem.
As a programmer, finding decent ODF libraries is far from certain. Last year I had to output some spreadsheets from a Go program, but I could not find any maintained library for ODS, so I had to output XLSX files. Recently, I was luckier while programming in Rust.
LibreOffice needs to pull their heads out of the sand and recognize that both OOXML and ODF are the past.
Make it a polyglot ZIP à la SingleFileZ if you have to (or polyglot JSON[1] or just straight up XHTML with a big sidecar blob of embedded metadata), but nothing trying to take on Microsoft Office is "the future" if it's trying to get there with a strategy shackled to the notion of people downloading the appropriate format-compatible software for something as simple as being able to view (not even edit!) the document that has been sent to them.
It's great to have open standards and free software, but be compatible by default with the universal formats understood by the readers for the ubiquitous document infrastructure that everyone already has installed (WHATWG-/W3C-compatible hypertext readers, i.e. Web browsers), or be forever doomed to the same level of obscurity that OpenOffice/LibreOffice were fated 10–15 years ago due to their myopia.
> A forward-looking format is one that reduces future dependency, not one that reinforces it[…] A “backward-looking” format, by contrast, is one that ties the future to the commercial strategies of a single vendor. In this sense, OOXML Transitional is an archaeological artefact that preserves the past at the expense of the future.
This is delusional. As long as it's modeled on the same outdated paradigm of 80s- and 90s-style office suites (no matter which one), then any format whether it's an ISO standard or not is hopelessly attempting to preserve the past.
1. <https://news.ycombinator.com/item?id=44229684>
Even today, web formats cannot render documents with the same fidelity as ODF. Especially not for spreadsheets.
Remember, many people feel that neither ODF or OOXML are sufficient and reach for things like LaTeX.
I think LibreOffice also supporting Markdown is a nice nod at simpler open formats.
Perhaps you have not dug into the problem deeply enough.
ODF is certainly better than OOXML at any rate.
> Even today, web formats cannot render documents with the same fidelity as ODF. Especially not for spreadsheets.
That's a nonsense claim—literally a category error. You're conflating the limitations of what can be represented in a given file format with an implementation's (lack of) support for doing things encoded in that format.
LibreOffice doesn't render documents encoded in plain-text, browser-compatible formats with the same fidelity as ODF because LibreOffice has not implemented encoding documents in plain-text, browser-compatible formats that it renders with the same fidelity as ODF.
There is nothing in ODF that's inherently unrepresentable in JSON or pure XML.
I am not sure what you claim about web formats.
If you try to print any non-trivial Web page with any of the existing Web browsers, all of them will print different things and none of them will print the same thing that they render on the computer display. For all of them the printed files will have various obvious defects, caused by incorrect size ratios between various elements or by an incorrect stacking of the elements, so that some of them obscure others that should have been visible.
For instance, Firefox and Chrome almost always print garbage, while Vivaldi is typically much better, but it also fails from time to time. It looks like all those who maintain Web browsers only test how pages are rendered on displays, but they never test how they are rendered by a print command.
With a file format intended for documents, the first property that I demand is that the document will be rendered perfectly in a deterministic way and it will look absolutely the same regardless on what medium it is rendered.
Nothing could be less true for anything that is used on the Web.
I strongly hate anyone who provides documentation in the form of Web pages, instead of using real documents, e.g. PDF files, ODF files or even Microsoft Office files, which can be used offline without problems. Nowadays, even the attempt to save Web pages in browsers is unreliable, due to the embedded scripts that may fail to work offline, making impossible the rendering of the saved pages.
> I am not sure what you claim about web formats.¶ If you try to print any non-trivial Web page with any of the existing Web browsers[…] For instance, Firefox and Chrome[…]
... why do you think this matters? Like, at all? We're talking about LibreOffice.
> Nothing could be less true for anything that is used on the Web.
This is, uh, crazy. In the first place, PDF is used on the Web, and it's both better suited for print-perfect sources and it's more portable. In the second place, believing that there's some magic inside ODF (or some other office suite's native format) that can't be represented in XML, JSON, or some other plain-text format is delusional.
If you claim that some text format used on the Web is suitable for storing a document, you should be able to point to some application that demonstrates the use of that text format in a manner that can be used for a document.
If none of the existing Web browsers can render a Web page in a predictable way in all circumstances, that clearly disqualifies all "Web formats". If the most complex programs like the Web browsers cannot always render correctly a Web page, then who can?
> There is nothing in ODF that's inherently unrepresentable in JSON or pure XML.
Since ODF is XML, that is a tautology.
I do not doubt that it is likely that ODF is more complicated than really necessary, and that something better could be designed.
However anyone who criticizes it should define an alternative better format and describe its advantages.
Pointing to some of the existing markup text formats is not this, because those lack a great number of features that are absolutely necessary for any document file format, so they are not comparable at all.
Saying that a markup text format can be suitable as a document format resembles the claim of the Americans that ASCII should be good enough for the speakers of non-English languages.
It would be great if you would stop and at least try to understand the argument you find yourself in and what's being claimed/described by the other side, especially in the comments you're choosing to respond to, instead of hallucinating some altogether completely different claims, like whether any two Web browsers consistently render HTML- and CSS-based Web pages in the wild the same—as if that's somehow pertinent.
> If none of the existing Web browsers can render a Web page in a predictable way in all circumstances, that clearly disqualifies all "Web formats"[†]. If the most complex programs like the Web browsers cannot always render correctly a Web page, then who can?
There is no brilliant observation here. We are not talking about how "the existing Web browsers can render a Web page in a predictable way in all circumstances". The answer to your question literally doesn't matter.
We're talking about the choices available to LibreOffice—choices about how to encode documents in a bytestream that can still be processed by apps that don't have explicit support for formats like ODF, and processed by those that do (apps like, you know, LibreOffice) that's (a) done in a way that's still consistent with the way that it (again: LibreOffice) treated the content the first time around, before it wrote it out to disk (b) consistent with the way that other apps that commit to the same standard of interoperability treat that content. Those are the _only_ tests that matter in this conversation. You can imagine all sorts of other tests and cross-examine against them, but no matter what you manage to come up with, they do not matter.
It's your insistence on not understanding this that's the problem.
The questions—again, the _only_ questions—to answer are:
1. Can LibreOffice encode its documents in such formats and in such a way that it doesn't compromise its ability to have the documents cleanly "round tripping" their way out (of LibreOffice) and back in (to LibreOffice) again—i.e., whether it can use one of these formats as its container (rather than the ZIP-for-XML-&c container that ODF uses). The answer to that question is an unequivocal "yes"—there is literally _nothing_ magical about ODF that can't be captured/encoded using some other scheme for representing data. It's equal parts baffling and maddening that this needs to be stated so, so, so, so, so many times in this discussion.
2. And whether LibreOffice—and any other desktop publishing app that aims to be interoperable—can render those documents "in a [reliably] predictable way in all circumstances". The answer is "yes—assuming it's able to do that now with ODF, and you aren't just putting your thumb on the scale in two dimensions [instead of just the one, as you've been doing throughout this exchange]".
Any and all other of these asinine tests and complaints about the deficient behavior of non-LibreOffice apps (like Firefox, or Chrome, or Windows 95's calc.exe…) are all half-thought-out red herrings that have next to no bearing on the topic at hand. Just mindnumbing attempts to move the goalposts.
† As I already pointed out in my previous comment, and which you inexplicably ignored, PDF is, in fact, quite good at this—and not even just that; it's so much better suited for what you're kvetching about.
> ODF is XML
Uh, no. ISO 26300[1], much like the Open Packaging Conventions[2] used for Microsoft's OOXML and XPS, "uses a package file to store the XML content of a document together with its associated binary data, and to optionally compress the XML content. This package is a standard Zip file". Leaving that aside and the monumental difference that that detail makes wrt the claim and its relevance to this situation:
There are no provisions within the spec requiring ODF producers make sure to spit out a ZIP bytestream that is carefully constructed so it can be processed by a pipeline that is expected to yield a meaningful result when it's treated as something other than ODF-flavored-XML-inside-ZIP (like application/pdf, image/png, image/svg+xml, or text/html, etc)—i.e. ODF packaging has none of the guarantees that SingleFileZ archives provide.
> However anyone who criticizes it should define an alternative better format and describe its advantages.
Boy am I done with this thread.
1. <https://www.iso.org/standard/66363.html>
2. <https://en.wikipedia.org/wiki/Open_Packaging_Conventions>
The media (web or desktop) is irrelevant: a file format must exists for backup and interoperability. I barely use office documents myself, but I work on software that produce and parse many spreadsheets every day.
An open standard is even more very relevant in public administrations where the process follows legal constraints and ISO standards. The Document Foundation's article reacts to an German institutional decision.
[dead]
.md is the future. I've compiled the worst word docs to markdown and then pdf, and suddenly it looks as great as TeX
None of the documents that I write for myself and much less any of the documents that I write for others could be written as .md.
The reason is that I always use only good typefaces and I must use various typeface features provided by the OTF format, to which something like .md does not provide access.
If .md would be extended with the missing capabilities, then it is likely that it would become not much simpler than ODF.
Even in the simplest documents that I write for myself, I must use multiple document styles, as provided by a format like ODF, because I typically mix multiple languages, including multiple alphabets.
.md is OK for a ReadMe file written in English, but it is not good enough for what I would call a real text document.
Just last week I needed to output a formatted document, and my choices were constrained to: HTML+CSS, Markdown, docx (OOXML), or odt (ODF).
The formatting had to be manipulable in code; the first two options were unsuitable for my use case. That left OOXML and ODF.
Both docx and odt are zip files, with XML content. Any guess which one turned out to be the better choice? I was left wondering why docx even exists in the first place.
Anyone using OOXML can rightly be regarded as stupid. There are far superior formats, and we should insist on their usage.
At some point mandating a universal document format for ALL government use, becomes the only reasonable way forwards, along side tax software, utility billing, employee payrole,recipts, and banking that is also standerdised, and free. Criminalising any further attempt to "patent" , obfusicate, privatise, or close source a public comunication medium as a shot accross the bow is the first step, with an expliset declaration that total dismantaling of companys attempting to insert themselves and comodify basic rights and privlages is going to happen. Maps and navigation are also on the list. Historical precidents are impossible to dismiss.