> The two office suites take very different paths here. LibreOffice uses the OpenDocument Format (ODF), an open standard meant to be controlled by no single company. Microsoft, on the other hand, created its own Office Open XML (OOXML) to support every feature in its own software, giving us the familiar .docx and .xlsx
It's so impressively underhandedly sneaky that Microsoft named their ODF-competitor format “Office Open” just as OpenOffice.org's (LibreOffice's direct ancestor) hype peaked with OO.o 2.0 having ODF as its native format, when MS Office finally had a viable and popular competitor for like the first time ever.
I highly doubt anyone except extremely technically literate individuals even heard of "Office Open XML" or "OOXML". To the average person, .docx is just the newer .doc
I think you are mixing up some things. There are no "likeword95" attributes.
At one point during standardization there was a proposal to add several attributes with names like that although most of them referred to non-Microsoft products like WordPerfect, but it didn't make it into the final standard. That's what you are probably thinking of.
Their purpose was to allow someone writing say a WordPerfect to OOXML converter to mark in the OOXML places that were using some specific WordPerfect formatting that couldn't be replicated in OOXML.
OOXML word processors were supposed to preserve those markings if they encountered them but never add them.
they dont care enough. I have said no to work that would require me to work with loads of MS shit, I have said no to open photoshop files, various other proprietary crap files.
They might care, but they just only care a tiny tiny tiny tiny bit, and certainly not enough to inconvenience themselves in any way for it.
I feel qualified to opine on this as both a former power user of Word and someone building a word processor for lawyers from scratch[1]. I've spent hours pouring over both the .doc and OOXML specs and implementing them. There's a pretty obvious journey visible in those specs from 1984 when computers were under powered with RAM rounding to zero through the 00's when XML was the hot idea to today when MSFT wants everyone on the cloud for life.
Unlike say an IDE or generic text editor where developers are excited to work on and dogfood the product via self-hosting, word processors are kind of boring and require separate testing/QA.
MSFT has the deep pockets to fund that development and testing/QA. LibreOffice doesn't.
The business model is just screaming that GPL'd LibreOffice is toast.
One sentence that stood out to me from there was this:
> The bottom line is that there are thousands of developer years of work that went into the current versions of Word and Excel, and if you really want to clone those applications completely, you’re going to have to do thousands of years of work.
This is blatantly not true. Only a small portion of all those "thousands of developer-years" is going to be actively present in these products at whatever point in time, as a lot of those developer-years are spent on replacing the output of other developer-years.
It's the difference between 117 billion humans ever having lived, and 8 billion humans currently living (and just some number of millions at any point in time before the industrial revolution - we've been around for a while, supposedly).
And this is still ignoring that someone looking to reimplement Office would be racing towards something pre-existing, rather than trying to come up with it in the first place. A lot of those developer-hours were spent on design and research, rather than rote implementation.
I half agree. The grow of the final version is not linear, but I guess it's still like sqrt(time) or something. There is a lot of refactoring and backend changes, but also there are a lot of obscure "features" that accumuate.
(It would be nice to make a study about the number of LOCs in Chrome or other big open source project. LOC is not a perfect metric, but it's better than just guesing.)
Word has a compatibility configuration window with a lot of weird features, like (totally made up) add 3 pixels if a bullet list starts a new page because it was the default in WinWord 1.7. And there are like 10 of to mimic 10 versions and perhaps even other editors.
Also, a log time ago Word has a "feature" that converted automatically every acronym into a "Mini Card" or "Smart Card" or something with a dotted underline and a rectangle that appeared when you put the mouse over it. It was annoying. There is still code to show them, and perhaps even code to create them.
Microsoft may have the deep pockets, but there are Word documents that LibreOffice opens correctly, MS Word 2007 opens correctly, and MS Word 2024 doesn't.
As AI tools become more dominant, businesses are going to want their documents to be fully read by their AI in whatever format they are in. I wouldn’t be surprised to see a fight over all of this brewing in the next couple of years.
How is the complexity relevant? LibreOffice already provides 99% of needed tools for most users, except a possibility to reliably read proprietary Word formats.
Its been Microsoft's strategy since its formation to make a lot of proprietary technology when it moves into any space and do so in a way that locks customers in such that if and when it is no longer the top product the customers can't easily leave. They do this in every single product and market they operate in. Where they can't ultimately win they buy their competitor and integrate the product then slowly kill it.
It is important to remind people of this, because they imagine that MS is integrating open source projects like git, linux, and others for the goodness of their heart. It's well know that this is just step 1 of embrace, extend, and extinguish. Next step (underway) is to add many features that will work only under the MS ecosystem and finally declare those original tools as legacy that should not be allowed in corporations.
They even do that with ISO C. They claim that portable functions like strlen are deprecated and insecure, and their recommended replacements are MSVC-specific.
I don't think anyone is under the impression that Microsoft is integrating with projects like git or linux out of the goodness of their heart. They do it for the exact same reason anyone does - thats were users/customers are, and they want to make money from them. This isn't some evil conspiracy, it's just normal boring ways to build products for people.
Microsoft is never looking for normal, boring ways to build products. A software company does't get much ahead thinking like this. They're using their old and successful strategy of embracing, extending, and extinguishing.
Honestly, having worked on Excel at Microsoft (though pretty far from the file format and a long time after OOXML was introduced), I'm pretty sure that the structure of OOXML is convoluted because it was easier to align with the data structures used by the app.
. . . being a savvy businessman is a con man? There's loads to criticize about MS in the 80s and 90s, but buying DOS fair and square and then building an ecosystem around it was just a good business move. The stuff they got sued and almost broken up over is the sketchy part.
The point that he's a con man is that he signed a document with IBM to supply an OS when he had none. Of course in retrospect he was "smart" to go around and buy one, but in fact he was promising IBM something he didn't have and in an alternate universe he could be sued for that.
If I order a new car, it may well not exist yet. The deal is that there will be a car on hand at some agreed upon date. I don’t think Gates did anything more illegal than the analogous car deal. More risky, perhaps, but that’s another thing entirely!
It's unethical to sell someone something if there are hidden risks. You wouldn't knowingly order a car from a company that tells you they know how to make cars but doesn't actually have the experience.
Maybe IBM was informed and understood the risk, but from what I've read about it Gates was less than forthcoming about what Microsoft actually had.
IF you order a car from an unknown company, you might get lucky and it turn out to be Lamborghini in 1963 but it also might turn out to be Aptera Motors in 2004.
However, Microsoft knew that a small company named Seattle Computer Products had developed an operating system similar to CP/M called QDOS, for Quick-and-Dirty Operating System. Microsoft suggested to IBM that QDOS could work as the IBM PC’s operating system. IBM asked Microsoft to license and further develop the operating system, which led to the formal contract on November 6, 1980. After the contract was signed, in December 1980 Microsoft would license the QDOS operating system to begin development of the IBM PC version.
New business idea: selling small luxury chairs to big tech CEOs. Subscription service, if they don't pay a million per month, they lose their license to jump over their chair.
How did this make the HN homepage? There isn't even any news here. It is an argument about ~20-year-old XML file formats, at a time when file formats couldn't matter less?
On top of that, Office supports OpenDocument formats, just like LibreOffice supports Office formats.
Also, IME the Office XML file format is far better supported by third parties - countless apps read/write them. I have multiple apps installed that can read/write an Office file, but MS Office is the only app on my machine that opens OpenDocument.
This is an old problem that is still harming the user freedom and hinders the adoption of LibreOffice today, due to Microsoft's anti-competitive practices. Are you saying it's not worth discussing?
As a user, I'd make that argument for sure. The casual user simply wants the least friction, Microsoft doesn't care, and the FOSS developer space just has to live in this environment. As part of a broader argument to develop for and use FOSS, there's maybe something worth thinking about here, but Microsoft's not going to change its ways on this even if a dozen of us complain.
I was annoyed about the Chrome/FF thing recently (well, annoyed by Chrome, and annoyed as a casual user by Firefox's relative inabilities), so thought about what it'd take to make a new web browser. It's FAR too complicated; there is way too much to implement. HTML/JS/CSS are excessively complex, made for use cases ~nobody will run into, and you pretty much are required to implement everything Chrome does for compatibility. It's crazy Firefox even exists.
I decided instead, as an exercise, to implement an entirely new Internet built on CrypticWeb running the Mystic Beaver Protocol (MBP); surely you've heard of these, they are very big in my household. It's very simple; instead of forking the Internet to add even more garbage, we start from scratch. Instead of JS, we use waterwheel (.ww) files which patch the python script running on the stateful lodge server. I even ported the server to Micropython and can run it on a tiny ESP32C6 that fits on my pinky. The client's written in Python and interprets the simple JSON payloads the server sends over, to render it and interact with the server as needed.
It all works fine, but people are locked into this overcomplicated Bad/Legacy/Corporate/Devil Internet. big smh; wake up, sheeple!
Microsoft Office isn't even compatible with itself, deapite its bespoke file format. For example formulas in Excel not being portable between different languages.
It is a plague across the whole industry. The format in this case is highly influenced by how one corp designed their own products. Multiple document formats have this problem. But you can also find the exact same thing in PCI DSS and other standards. Like, one corp designed a tool to scan for a certain flaw and suddenly it is mandatory. Just ridiculous.
It’s not just the complex XML based format. Word has collaboration tie-in’s with Skype, Teams, SharePoint, OneDrive etc
It’s an entire ecosystem
Also, I have tried to use LibreOffice and you have to learn an entirely new tool. The user interfaces are different. Word has its own issues of course but LibreOffice does not feel as polished
There are things in Word that are legacy and carry overs from another time that carry various nuance. It’s not all documented set of features either
Trying to replicate the entire look and feel is incredibly difficult
Most people are going to encounter Word in a corporate setting and to have them switch to another tool is going to a big hill to climb
You're right for home users. In businesses, the hill is also that some users are power users that have locked themselves in, in slightly different ways between the different power users. Also the company has also locked itself in by drinking deeply from the Microsoft well (e.g. AD and sharepoint and Windows etc) and marketing away will cost them a lot of time and effort, and therefore money.
If the lawyers and the financial analysts need Windows Office for certain uses let them use them. But I'd note that the first edition of a book I wrote a while back had a Sharepoint workflow and had moved away by the second edition.
The licensing/support is cheaper carries some weight. But Windows to Linux mostly didn't win a lot of fights on the desktop. But Google Docs for collaboration and general simplicity does win over a lot of companies.
My comment was focused on Word. There is definitely a lot of lock-in for the larger MS ecosystem in business. But that is the nature of large bureaucracies. They struggle to switch between UPS and FedEx, or between Staples and OfficeDepot. Switching away from the MS ecosystem is significantly harder than that.
> I have tried to use LibreOffice and you have to learn an entirely new tool.
I use word processors so rarely that every time it's like learning a new tool. Whether it's Word, Google Doc, LibreOffice, or anything else.
I will say that Google Docs and Word both feel a bit more "polished" than LibreOffice which still feels very distinctly like a 1990's era desktop program. I guess because it is.
Latterly, I worked at a predominantly open source company that switched from LibreOffice to Google Docs over the screams of many. It was a night and day improvement both in terms of the software itself and the ability to collaborate.
Libreoffice was formerly openoffice which was formerly staroffice, an office suite that has its deepest origin in Starwriter, a word processor for the Zilog Z80 released in 1985.
At one point LibreOffice + MS Office were pretty much on par with each other.
But MS has built this giant moat of integrated proprietary services around these systems that make it difficult to switch away once you are sucked into the environment.
It takes a pretty sizable expense to switch to anything else, while satisfying all of a companies different workflows for various roles and levels of experience.
If not MS Office + it's M365 Eco system, what then? Google Workspace? That's kinda the same problem in a different color?
Google tried to get kids hooked in school and it’s decent. But when you want to do serious work, you need to use Microsoft products. Google’s product is like a toy
Most serious users can use things other than Microsoft just fine. The problem is file compatibility. Unfortunately, network effects are very much a thing. Plus most users just don't realize they don't actually have to keep shovelling money at Microsoft.
I don't know much about advanced Google Docs, but I know that's the far from true for Google Sheets. Google Sheets covers a shit ton of ground that Excel covers, but a serious user can never replace Excel with Sheets.
We had a lot of marketing and sales people who wanted to use Google Sheets instead of Excel to load sales reports from a shared PG database. This is of course a built-in feature (loading a PG table as a sheet) in Excel. Google Sheets obviously doesn't support any imports. They are all paid plugins, and they all make you pay per row or cell or column.
You can write a plugin to do it, except their funky AppScript (which is a custom synchronous Javascript/V8 environment where there is no async or callbacks, and promises block.
Except, the plugin can only connect to MySQL. PG support ask has been open since 2011. They decided in 2014 that they don't plan to support PG. Then in 2018 they thought maybe they can only support GCP Postgres.[1]
Ok, fine. You make your own API to call and load data. No, you can't return more than 50 MBs. So better split your query.
Oh, you want to load a 50MB CSV? There are 3 different APIs for that:
- "Import from GDrive". This is the one you want. Import 50MB in a couple of seconds. But it also requires the most permissions to full access the user's GDrive
- "Basic Import API". This imports at ~100kbps but not always. gets slower for large data
- "Advanced Import API". This imports at ~250kbps but not always. gets slower for small data
and you need to run your own benchmarks[2][3] to understand which API to use in which context, then keep reruning them as things change.
I am always amazed at how much people can bash Microsoft products when the whole reason to use them is often that they are the best thing around, with the most features and allows you to do things that would be hard/impossible otherwise.
The most egregious are the apple fanboys who try to pretend that the iWork suite is even close to being a replacement for Microsoft office.
Like sure, if you only need to write letters and resumes it will do but at this point any free office suite will do...
This doesn't make sense generically since not every software has all the features of other software.
But to be more specific - pick their concern, describe the replacement workflow with Sheets, then explain how it's a good replacement
> The only question is wether they want to learn doing things differently.
Of course it's not the only question, among many other questions could be whether they can afford to spend a lot of time/money after learning (though this investment also shouldn't be ignored) to do the same thing, and whether they can do the same for all the people depending on the current workflow
I have the opposite opinion. If I have something that I don’t care if it’s irrecoverable 5 years from now, or I don’t care that it prints differently on each printer, MS Office can fit the bill.
I collaborate with others, a lot, though. Google is simply better at that
I used Google Docs a lot initially for my Masters for our team projects but eventually we switched to our corporate Teams collaboration space. We couldn’t get Google Docs to match the formats needed for assignments. Word worked out of the box.
If LibreOffice were smart, they'd introduce free licensing for schools and universities, so the students could learn it and then ask for it when they get to jobs later on.
Maybe they could even release the source under a copyleft license, so the students can learn from it and maybe contribute.
What is important here is that when confronted with customer interest in the open ODF standard, Microsoft created an entirely different "open" standard and then used their influence to push it. This is a routine technique used by dominant players in many technical industries. The trick is to not push your standard too hard, then you risk that it becomes an actual industry standard. You need to maintain a balance to maintain both standards, thus significantly reducing the power of both.
Perhaps Microsoft is at the point where it doesn't feel all that threatened by ODF and is reducing the value of their standard with complexity. Does Microsoft even bother to produce documents that conform to either standard any more?
I mean… sure? When I saw this headline I was imagining that Microsoft added a brand-new ultracomplicated format. But no, the article is solely about OOXML. Why is the blog post re-litigating a fight that LibreOffice already fought almost 20 years ago?
Hasn't this always been an argument against it? I remember hearing the same arguments when we tried to get the EU to ditch OpenXML over a decade ago al least.
Everyone needs to stop competing with Office and just build a better office.
The truth is, no one needs to be compatible these days. Everyone is either:
- using the same software company-wide, whether it's LO, Google Docs or MS Office
- exporting to PDF when sharing docs with someone outside the organization
The real thing that LO is missing is server side hosting and easy syncing... Until they get that, it's just going to be something used by individuals and small orgs.
I still remember, at my university we had LO installed on lab computers, MS Office was a "requirement" for students to purchase, but most profs simply insisted we simply hand in assignments as Google Docs links because there was incompatibilities between .doc, .docx, the formats made by LO, etc... Google Docs were the only ones that could be shared and be 100% identical on every computer with the link.
If LO had a web hosted solution and provided easy to install server code for organizations, they'd dominate. But they don't...
Hmmm... They probably need to advertise that more, or connect the two. Had no idea they were related. Also the Collabora website is cancer, it's not easy at all trying it...
I tried really hard to use Collabora 'web' solution. It is unusable.
For those who don't know: it renders the whole LibreOffice interface on server and passes it to you. Lots of issues with hotkeys on non-standard layouts (I use Dvorak), with language switching, with mouse behavior, with clipboard, ugh. Of course it is hungry for resources. No. This isn't the way.
Are you sure that's not just a VNC-based demo of the desktop version of Collabora Office?
I wouldn't be too surprised if their web version was like that (given how long ago the first version came out), except Nextcloud uses a Collabora-based system for its online editing, and I doubt they'd go with something like that.
Never attribute malice where incompetence is a sufficient explanation.
Microsoft isn't intentionally obfuscating the docx. Docx is a shit-show because hundreds upon hundreds of Microsoft business initiatives, executive pet projects, and ancient compatibility rules have all collided to make a giant pile of dung.
If LibreOffice is worried about what docx does to their productivity ... you should see the fucking engineers lamenting INSIDE Microsoft about what it does to their friggin productivity.
This horseshit isn't anyone's plan. This horseshit is an emergent phenomenon like a fucking termite hill in your back yard ... where no single termite is responsible or knoweldgeable, but all of them together made a pile that breaks your lawnmower.
I doubt this one is explicit "locking in", as much as it's reflecting the increasing internal complexity and lack of focus on product quality. I'd be willing to bet the internal teams dread working with the overly complex structures too.
I know about the lack of tech-savvyness of most humans, but isn't Markdown and Pandoc—if you slam a GUI in front of it—covering the needs of 99% of users?
Granted, when you need formatting, like for a formal letter, you use a template someone made but this is not what most people use Word for.
And don't get me started on "people wouldn't understand how to put things in bold or italics"; they can barely use Word anyway. Might as well use something much simpler. Office "productivity" suites are over to me.
(1) These look great - I saved the Impacts so I can keep reading it later. Can you share what the background/reason is for this document?
(2) Hoping you might be able to point me in the right direction. I often take notes with clients in Markdown, and then after the call, do some cleanup. At that point, I would love to export it as a slick-looking PDF, and also HTML which I could paste into Apple Mail. I've done a few experiments with Markdown --> Pandoc --> HTML, but I started to realize I would need to dig deeper into Pandoc, figure out its HTML templates, create CSS. And wasn't sure if I ought to be doing Markdown-->HTML-->PDF. My ideal "flow" would be something like...
When digging around in Pandoc, I also got the impression I would need to learn Tex formatting. I haven't figured out a "stack" on this topic, so thought you might have ideas.
Currently, I am using Obsidian for taking notes. Then I use the "Copy as HTML" 3rd party plugin to copy my Markdown notes as basically themed HTML. I then open up MacOS TextEdit, on Rich Text mode, and paste that in. I might do a bit of text formatting here, clean out [ ] markdown links, etc. Next, I copy & paste from TextEdit to Apple Mail. I sometimes color the headings manually, and other basic formatting. Here I'll add the "Hello, " and "Sincerely" wrap around the notes, and send. I'll then check my iPhone Mail to make sure the mail looks OK on Dark Mode.
Seems like a huge gap in the "note taking" tooling, but I'm hoping that I'm just ignorant of a few solid tools that can handle this workflow: "Take notes on a call, send them nicely formatted to client in email / PDF"
cmark is an easy to build C project that reads markdown (commonmark, but close enough) that I also write in obsidian, and writes it as xml or html. You can probably automate some of that conversion process, e.g. with xslt, though I'll admit that's an acquired taste.
I can confirm. When trying convert simple Word sentences and tables to e.g. Markdown/HTML from a Word XML you need a PhD in XML edge cases and nested garbage.
I was amazed when I realised that Word docs were just zip files and you could poke around in the xml files embedded inside of them.
I almost implemented a working React -> Word document renderer back in 2017, but it didn't have support for creating the xml tags with : inside of them (which OOXML documents use).
Even though markitdown is a Microsoft project, it's just a thin wrapper around a bunch of 3rd party Python packages. For example, to go from docx to Markdown, it uses mammoth to convert docx to HTML[0], then uses markdownify to convert the HTML into Markdown[1].
Technically, they're a bit more than just zip files (they're OPC containers [0]), but if you're hand editing the file content it doesn't really matter.
Well, it is not pretty to see how the sausage gets made, but extracting formatted text from docx is absolutely doable, no PhD involved. Source: I have done it as a little sidequest because it was useful to audit a set of word documents.
I loved OpenOffice.org back in the day, but in today's world with no modern web collaboration option it is dead. I'm aware of the attempts to make it run via browser, and no, they aren't really there, not even close.
What we need today is a web-first suite of apps that does everything Google Docs/Spreadsheets/Slides do, but uses OpenDocument family of standards as their native file format.
It isn't dead because it can still be used for many use cases. I write many documents in LibreOffice Writer and I regularly use LibreoOffice Calc for financial calculations. Also, LibreOffice Calc is a stupid name because it conflicts with the calculator in application launchers that search for substrings.
I use Libreoffice at work to open and edit MS Office files and it works mostly fine, except for PowerPoint (missing fonts but also general madness). In general, I think the LibreOffice folks did a great job, esp. when it comes to MS file format support. And of course they need to, because I (and I guess most people) actually open more Ms files with LibreOffice than files in LOs native formats.
After having worked for many companies, I think the complex file formats is likely due to employee turnover, pressure from management to just get shit done, and careless coding. Not ill intended anti-competitive practices. The locking in is likely an unintended bonus for Microsoft.
I didn’t realise we’d entered a time machine to 2007. I’ve worked extensively with OOXML and yes, the documentation is cryptic and often absent, but Microsoft will help you out if you contact them on their forums. I see Libre Office devs there all the time!
But the complexity is not some kind of conspiracy - it’s inherent - it comes from the fact that Office is ancient and very, very complex with a huge number of features. Many features are implemented in backwards compatible way on top of the old version of similar features and then the whole thing has been back ported from a bunch of C structure to XML which has the most woeful and underpowered schema language imaginable.
> it comes from the fact that Office is ancient and very, very complex with a huge number of features.
IIRC one of the many unfortunate decisions made by MS with OOXML (whether intentionally, or not, or both) is to codify a lot of display and formatting quirks directly in the schema with very little explanation or docs. Instead of making it s different namespace or layer.
So, to implement OOXML, you also needed to reverse engineer, say, behavior of Word97 etc.
The PDF spec is well designed to be both very expandable and easily readable.
It's absolutely horrible to try and edit.
That's because the structure of a PDF is essentially a bunch of media "streams". It's very easy to say "render a jpeg at this location on the page" but that's about it. It doesn't store, for example, the fact that you might need to wrap words around a page. Instead, it's "Here's a box with text in it".
The only thing that really could make PDF rendering hard is adobe put a whole bunch of garbage into the spec. For example, the full spec had the ability to run javascript and flash at one point (not sure if it does anymore).
No. There are hundreds of programs that easily read and create PDFs. OTOH, reading .docx is a pain. Far be it from me to defend Adobe, but PDF is nothing like MS Office formats.
Yeah it's a great app, one of the big reasons I still use MacOS - but there's certain features which you still need Adobe Acrobat for, I know to fill in the tax forms for my country you do.
groklaw.net dived deep into this.
AFAICR, MS never achieved round tripping between their own formats, let alone ODF. It was a lock in scheme where MS would delegate people to the ODF committees (don't remember the official name) to sabotage them from the inside.
You might not be wrong, but yelling at people for accepting conveniences that most don't even understand or care about, seems a bit pointless to me. Why even stop at document formats? Why not outrage over their choice of Windows itself?
What good do you think this does? I'm genuinely curious.
> Why not outrage over their choice of Windows itself?
Because it doesn't necessarily affect anyone. Using Windows ipso facto doesn't mean you will send someone a file they can't read without a Microsoft program.
I have two Windows machines in my home; they have LibreOffice on them, as well as Firefox.
No, but I think the people most likely to be oblivious to all this, will be the ones using Microsoft programs that write in proprietary formats by default... and I would say they're not even wrong if that's what most other people use too.
People send me Outlook messages all the time that I can't read at all, and many Excel spreadsheets and Powerpoint presentations with features that aren't supported in anything on Linux. I literally have to use a Windows VM with proper Office to read this stuff. And not only work documents, but documents from my children's school/teachers as well.
> The two office suites take very different paths here. LibreOffice uses the OpenDocument Format (ODF), an open standard meant to be controlled by no single company. Microsoft, on the other hand, created its own Office Open XML (OOXML) to support every feature in its own software, giving us the familiar .docx and .xlsx
It's so impressively underhandedly sneaky that Microsoft named their ODF-competitor format “Office Open” just as OpenOffice.org's (LibreOffice's direct ancestor) hype peaked with OO.o 2.0 having ODF as its native format, when MS Office finally had a viable and popular competitor for like the first time ever.
https://www.openoffice.org/press/2.0/press_release.html (2005-10-20)
https://news.microsoft.com/2005/11/21/qa-microsoft-co-sponso... (2005-11-21)
I highly doubt anyone except extremely technically literate individuals even heard of "Office Open XML" or "OOXML". To the average person, .docx is just the newer .doc
and ooxml is basically just a serialization of microsoft office internals, with attributes such as "likeword95" on certain elements..
microsoft made a total piece of steaming turd, and its users dont care.
I think you are mixing up some things. There are no "likeword95" attributes.
At one point during standardization there was a proposal to add several attributes with names like that although most of them referred to non-Microsoft products like WordPerfect, but it didn't make it into the final standard. That's what you are probably thinking of.
Their purpose was to allow someone writing say a WordPerfect to OOXML converter to mark in the OOXML places that were using some specific WordPerfect formatting that couldn't be replicated in OOXML.
OOXML word processors were supposed to preserve those markings if they encountered them but never add them.
There was/is a <autoSpaceLikeWord95>
https://learn.microsoft.com/en-us/dotnet/api/documentformat....
Huh. Looks like some of them came back. Here’s one of the WordPerfect ones.
https://learn.microsoft.com/en-us/dotnet/api/documentformat....
I wonder if there is a complete list readily available?
Yes, it's bad, but the users _do_ care. Most don't have a choice, business machines have MS Office/Outlook/Teams as default and they can't change.
they dont care enough. I have said no to work that would require me to work with loads of MS shit, I have said no to open photoshop files, various other proprietary crap files.
They might care, but they just only care a tiny tiny tiny tiny bit, and certainly not enough to inconvenience themselves in any way for it.
I feel qualified to opine on this as both a former power user of Word and someone building a word processor for lawyers from scratch[1]. I've spent hours pouring over both the .doc and OOXML specs and implementing them. There's a pretty obvious journey visible in those specs from 1984 when computers were under powered with RAM rounding to zero through the 00's when XML was the hot idea to today when MSFT wants everyone on the cloud for life.
Unlike say an IDE or generic text editor where developers are excited to work on and dogfood the product via self-hosting, word processors are kind of boring and require separate testing/QA.
MSFT has the deep pockets to fund that development and testing/QA. LibreOffice doesn't.
The business model is just screaming that GPL'd LibreOffice is toast.
[1] Plug: https://tritium.legal
There is an old post by Joel Spolsky that worked as PM in Excel a looong time ago, and he agree with you: "Why are the Microsoft Office file formats so complicated? (And some workarounds)" https://www.joelonsoftware.com/2008/02/19/why-are-the-micros... (HN discussions https://news.ycombinator.com/item?id=12471604 (393 points | Sept 2016 | 229 comments) and https://news.ycombinator.com/item?id=118909 (60 points | Feb 2008 | 20 comments))
One sentence that stood out to me from there was this:
> The bottom line is that there are thousands of developer years of work that went into the current versions of Word and Excel, and if you really want to clone those applications completely, you’re going to have to do thousands of years of work.
This is blatantly not true. Only a small portion of all those "thousands of developer-years" is going to be actively present in these products at whatever point in time, as a lot of those developer-years are spent on replacing the output of other developer-years.
It's the difference between 117 billion humans ever having lived, and 8 billion humans currently living (and just some number of millions at any point in time before the industrial revolution - we've been around for a while, supposedly).
And this is still ignoring that someone looking to reimplement Office would be racing towards something pre-existing, rather than trying to come up with it in the first place. A lot of those developer-hours were spent on design and research, rather than rote implementation.
I half agree. The grow of the final version is not linear, but I guess it's still like sqrt(time) or something. There is a lot of refactoring and backend changes, but also there are a lot of obscure "features" that accumuate.
(It would be nice to make a study about the number of LOCs in Chrome or other big open source project. LOC is not a perfect metric, but it's better than just guesing.)
Word has a compatibility configuration window with a lot of weird features, like (totally made up) add 3 pixels if a bullet list starts a new page because it was the default in WinWord 1.7. And there are like 10 of to mimic 10 versions and perhaps even other editors.
Also, a log time ago Word has a "feature" that converted automatically every acronym into a "Mini Card" or "Smart Card" or something with a dotted underline and a rectangle that appeared when you put the mouse over it. It was annoying. There is still code to show them, and perhaps even code to create them.
Microsoft may have the deep pockets, but there are Word documents that LibreOffice opens correctly, MS Word 2007 opens correctly, and MS Word 2024 doesn't.
> The business model is just screaming that GPL'd LibreOffice is toast.
Only if Word formats remain dominant. There might be hope with the EU moving off Word that an alternative, real standard might take root.
As AI tools become more dominant, businesses are going to want their documents to be fully read by their AI in whatever format they are in. I wouldn’t be surprised to see a fight over all of this brewing in the next couple of years.
The document formats are largely irrelevant. The problem is the editor itself, and its enormous complexity.
How is the complexity relevant? LibreOffice already provides 99% of needed tools for most users, except a possibility to reliably read proprietary Word formats.
*poring over
Lawyers also tend to pore a lot, so it's worth getting the word right! ;-)
There are non-native English speaking lawyers in the world afaik
TIL
Its been Microsoft's strategy since its formation to make a lot of proprietary technology when it moves into any space and do so in a way that locks customers in such that if and when it is no longer the top product the customers can't easily leave. They do this in every single product and market they operate in. Where they can't ultimately win they buy their competitor and integrate the product then slowly kill it.
It is important to remind people of this, because they imagine that MS is integrating open source projects like git, linux, and others for the goodness of their heart. It's well know that this is just step 1 of embrace, extend, and extinguish. Next step (underway) is to add many features that will work only under the MS ecosystem and finally declare those original tools as legacy that should not be allowed in corporations.
They even do that with ISO C. They claim that portable functions like strlen are deprecated and insecure, and their recommended replacements are MSVC-specific.
I don't think anyone is under the impression that Microsoft is integrating with projects like git or linux out of the goodness of their heart. They do it for the exact same reason anyone does - thats were users/customers are, and they want to make money from them. This isn't some evil conspiracy, it's just normal boring ways to build products for people.
> just normal boring ways to build products
Microsoft is never looking for normal, boring ways to build products. A software company does't get much ahead thinking like this. They're using their old and successful strategy of embracing, extending, and extinguishing.
A little from column A, a little from column B.
Honestly, having worked on Excel at Microsoft (though pretty far from the file format and a long time after OOXML was introduced), I'm pretty sure that the structure of OOXML is convoluted because it was easier to align with the data structures used by the app.
My take as an interested third party as well.
Yah and they’ve only gotten sneakier with it. VS Code and the proprietary Pylance, remote ssh, etc
Gates never even made DOS. He bought it from someone else and rebranded it. He's been a con man since day 1.
. . . being a savvy businessman is a con man? There's loads to criticize about MS in the 80s and 90s, but buying DOS fair and square and then building an ecosystem around it was just a good business move. The stuff they got sued and almost broken up over is the sketchy part.
The point that he's a con man is that he signed a document with IBM to supply an OS when he had none. Of course in retrospect he was "smart" to go around and buy one, but in fact he was promising IBM something he didn't have and in an alternate universe he could be sued for that.
If I order a new car, it may well not exist yet. The deal is that there will be a car on hand at some agreed upon date. I don’t think Gates did anything more illegal than the analogous car deal. More risky, perhaps, but that’s another thing entirely!
It's unethical to sell someone something if there are hidden risks. You wouldn't knowingly order a car from a company that tells you they know how to make cars but doesn't actually have the experience.
Maybe IBM was informed and understood the risk, but from what I've read about it Gates was less than forthcoming about what Microsoft actually had.
IF you order a car from an unknown company, you might get lucky and it turn out to be Lamborghini in 1963 but it also might turn out to be Aptera Motors in 2004.
Based on the top links from a quick search, it seems IBM was fully aware that they were sub-licensing the OS through Microsoft:
https://www.pcmag.com/news/the-rise-of-dos-how-microsoft-got...
https://thisdayintechhistory.com/11/06/ibm-signs-a-deal-with...
From the second link:
However, Microsoft knew that a small company named Seattle Computer Products had developed an operating system similar to CP/M called QDOS, for Quick-and-Dirty Operating System. Microsoft suggested to IBM that QDOS could work as the IBM PC’s operating system. IBM asked Microsoft to license and further develop the operating system, which led to the formal contract on November 6, 1980. After the contract was signed, in December 1980 Microsoft would license the QDOS operating system to begin development of the IBM PC version.
Still the only big tech CEO who can jump over a chair tho
New business idea: selling small luxury chairs to big tech CEOs. Subscription service, if they don't pay a million per month, they lose their license to jump over their chair.
But I bet he can’t throw them the way monkey boy Ballmer did.
How did this make the HN homepage? There isn't even any news here. It is an argument about ~20-year-old XML file formats, at a time when file formats couldn't matter less?
On top of that, Office supports OpenDocument formats, just like LibreOffice supports Office formats.
Also, IME the Office XML file format is far better supported by third parties - countless apps read/write them. I have multiple apps installed that can read/write an Office file, but MS Office is the only app on my machine that opens OpenDocument.
This is an old problem that is still harming the user freedom and hinders the adoption of LibreOffice today, due to Microsoft's anti-competitive practices. Are you saying it's not worth discussing?
As a user, I'd make that argument for sure. The casual user simply wants the least friction, Microsoft doesn't care, and the FOSS developer space just has to live in this environment. As part of a broader argument to develop for and use FOSS, there's maybe something worth thinking about here, but Microsoft's not going to change its ways on this even if a dozen of us complain.
I was annoyed about the Chrome/FF thing recently (well, annoyed by Chrome, and annoyed as a casual user by Firefox's relative inabilities), so thought about what it'd take to make a new web browser. It's FAR too complicated; there is way too much to implement. HTML/JS/CSS are excessively complex, made for use cases ~nobody will run into, and you pretty much are required to implement everything Chrome does for compatibility. It's crazy Firefox even exists.
I decided instead, as an exercise, to implement an entirely new Internet built on CrypticWeb running the Mystic Beaver Protocol (MBP); surely you've heard of these, they are very big in my household. It's very simple; instead of forking the Internet to add even more garbage, we start from scratch. Instead of JS, we use waterwheel (.ww) files which patch the python script running on the stateful lodge server. I even ported the server to Micropython and can run it on a tiny ESP32C6 that fits on my pinky. The client's written in Python and interprets the simple JSON payloads the server sends over, to render it and interact with the server as needed.
It all works fine, but people are locked into this overcomplicated Bad/Legacy/Corporate/Devil Internet. big smh; wake up, sheeple!
[dead]
Microsoft Office isn't even compatible with itself, deapite its bespoke file format. For example formulas in Excel not being portable between different languages.
Yeah this is basically just an ad for LibreOffice. To be complaining about XML in Office in 2025 is wild.
Why does the title say "slams" but neither the headline or URL contain "slams"? I think anything that says slams is not worth reading.
Saving 4 characters (over "calls out") to fit within the HN submission title length limit, I'd assume.
I don't think microsoft could feel a slam from anyone tbh, much less libreoffice.
it's ridiculous, clearly to get eyeballs. there should be a clickbait label on such titles
Headlinese
It is a plague across the whole industry. The format in this case is highly influenced by how one corp designed their own products. Multiple document formats have this problem. But you can also find the exact same thing in PCI DSS and other standards. Like, one corp designed a tool to scan for a certain flaw and suddenly it is mandatory. Just ridiculous.
It’s not just the complex XML based format. Word has collaboration tie-in’s with Skype, Teams, SharePoint, OneDrive etc
It’s an entire ecosystem
Also, I have tried to use LibreOffice and you have to learn an entirely new tool. The user interfaces are different. Word has its own issues of course but LibreOffice does not feel as polished
There are things in Word that are legacy and carry overs from another time that carry various nuance. It’s not all documented set of features either
Trying to replicate the entire look and feel is incredibly difficult
Most people are going to encounter Word in a corporate setting and to have them switch to another tool is going to a big hill to climb
I have met very few non-techies that could tell the difference between Word and Wordpad or use them any differently.
Most people below the age of 30 can switch between Google Docs or Word without blinking. They don't use more than a few of the features of either.
This "big hill" you mention is a fantasy.
You're right for home users. In businesses, the hill is also that some users are power users that have locked themselves in, in slightly different ways between the different power users. Also the company has also locked itself in by drinking deeply from the Microsoft well (e.g. AD and sharepoint and Windows etc) and marketing away will cost them a lot of time and effort, and therefore money.
If the lawyers and the financial analysts need Windows Office for certain uses let them use them. But I'd note that the first edition of a book I wrote a while back had a Sharepoint workflow and had moved away by the second edition.
The licensing/support is cheaper carries some weight. But Windows to Linux mostly didn't win a lot of fights on the desktop. But Google Docs for collaboration and general simplicity does win over a lot of companies.
My comment was focused on Word. There is definitely a lot of lock-in for the larger MS ecosystem in business. But that is the nature of large bureaucracies. They struggle to switch between UPS and FedEx, or between Staples and OfficeDepot. Switching away from the MS ecosystem is significantly harder than that.
> I have tried to use LibreOffice and you have to learn an entirely new tool.
I use word processors so rarely that every time it's like learning a new tool. Whether it's Word, Google Doc, LibreOffice, or anything else.
I will say that Google Docs and Word both feel a bit more "polished" than LibreOffice which still feels very distinctly like a 1990's era desktop program. I guess because it is.
Latterly, I worked at a predominantly open source company that switched from LibreOffice to Google Docs over the screams of many. It was a night and day improvement both in terms of the software itself and the ability to collaborate.
LibreOffice very much took cues from the old Office setup. While alternatives like OnlyOffice took cues from the current Office setup.
Libreoffice was formerly openoffice which was formerly staroffice, an office suite that has its deepest origin in Starwriter, a word processor for the Zilog Z80 released in 1985.
At one point LibreOffice + MS Office were pretty much on par with each other.
But MS has built this giant moat of integrated proprietary services around these systems that make it difficult to switch away once you are sucked into the environment.
It takes a pretty sizable expense to switch to anything else, while satisfying all of a companies different workflows for various roles and levels of experience.
If not MS Office + it's M365 Eco system, what then? Google Workspace? That's kinda the same problem in a different color?
Google tried to get kids hooked in school and it’s decent. But when you want to do serious work, you need to use Microsoft products. Google’s product is like a toy
Most serious users can use things other than Microsoft just fine. The problem is file compatibility. Unfortunately, network effects are very much a thing. Plus most users just don't realize they don't actually have to keep shovelling money at Microsoft.
I don't know much about advanced Google Docs, but I know that's the far from true for Google Sheets. Google Sheets covers a shit ton of ground that Excel covers, but a serious user can never replace Excel with Sheets.
We had a lot of marketing and sales people who wanted to use Google Sheets instead of Excel to load sales reports from a shared PG database. This is of course a built-in feature (loading a PG table as a sheet) in Excel. Google Sheets obviously doesn't support any imports. They are all paid plugins, and they all make you pay per row or cell or column.
You can write a plugin to do it, except their funky AppScript (which is a custom synchronous Javascript/V8 environment where there is no async or callbacks, and promises block.
Except, the plugin can only connect to MySQL. PG support ask has been open since 2011. They decided in 2014 that they don't plan to support PG. Then in 2018 they thought maybe they can only support GCP Postgres.[1]
Ok, fine. You make your own API to call and load data. No, you can't return more than 50 MBs. So better split your query.
Oh, you want to load a 50MB CSV? There are 3 different APIs for that:
- "Import from GDrive". This is the one you want. Import 50MB in a couple of seconds. But it also requires the most permissions to full access the user's GDrive
- "Basic Import API". This imports at ~100kbps but not always. gets slower for large data
- "Advanced Import API". This imports at ~250kbps but not always. gets slower for small data
and you need to run your own benchmarks[2][3] to understand which API to use in which context, then keep reruning them as things change.
[1]: https://issuetracker.google.com/issues/36752790?pli=1
[2]: https://gist.github.com/tanaikech/d102c9600ba12a162c667287d2...
[3]: https://gist.github.com/tanaikech/030203c695b308606041587e6d...
I am always amazed at how much people can bash Microsoft products when the whole reason to use them is often that they are the best thing around, with the most features and allows you to do things that would be hard/impossible otherwise.
The most egregious are the apple fanboys who try to pretend that the iWork suite is even close to being a replacement for Microsoft office. Like sure, if you only need to write letters and resumes it will do but at this point any free office suite will do...
> but a serious user can never replace Excel with Sheets.
This doesn't mean anything.
If you read further, you might see understand of the meaning
There are stuff "serious user" do they would do differently with other software. Nothing is irreplaceable.
The only question is wether they want to learn doing things differently.
This doesn't make sense generically since not every software has all the features of other software.
But to be more specific - pick their concern, describe the replacement workflow with Sheets, then explain how it's a good replacement
> The only question is wether they want to learn doing things differently.
Of course it's not the only question, among many other questions could be whether they can afford to spend a lot of time/money after learning (though this investment also shouldn't be ignored) to do the same thing, and whether they can do the same for all the people depending on the current workflow
If you repeat it long enough, it'll be true.
I have the opposite opinion. If I have something that I don’t care if it’s irrecoverable 5 years from now, or I don’t care that it prints differently on each printer, MS Office can fit the bill.
I collaborate with others, a lot, though. Google is simply better at that
I used Google Docs a lot initially for my Masters for our team projects but eventually we switched to our corporate Teams collaboration space. We couldn’t get Google Docs to match the formats needed for assignments. Word worked out of the box.
What was “the format needed for the assignment”? Was it “Microsoft word”?
The content and format of the text needed could not be replicated in Google Docs like the proper fonts, headers, footers etc
If LibreOffice were smart, they'd introduce free licensing for schools and universities, so the students could learn it and then ask for it when they get to jobs later on.
Maybe they could even release the source under a copyleft license, so the students can learn from it and maybe contribute.
LibreOffice is free and GPL.
That's the joke, yes. Though it's MPL, not GPL: https://www.libreoffice.org/about-us/licenses
Woosh!
What is important here is that when confronted with customer interest in the open ODF standard, Microsoft created an entirely different "open" standard and then used their influence to push it. This is a routine technique used by dominant players in many technical industries. The trick is to not push your standard too hard, then you risk that it becomes an actual industry standard. You need to maintain a balance to maintain both standards, thus significantly reducing the power of both.
Perhaps Microsoft is at the point where it doesn't feel all that threatened by ODF and is reducing the value of their standard with complexity. Does Microsoft even bother to produce documents that conform to either standard any more?
Original blog post by LibreOffice is here: https://blog.documentfoundation.org/blog/2025/07/18/artifici...
I mean… sure? When I saw this headline I was imagining that Microsoft added a brand-new ultracomplicated format. But no, the article is solely about OOXML. Why is the blog post re-litigating a fight that LibreOffice already fought almost 20 years ago?
Because it still matterts today?
Those who forget the past..?
Hasn't this always been an argument against it? I remember hearing the same arguments when we tried to get the EU to ditch OpenXML over a decade ago al least.
[0] https://interoperable-europe.ec.europa.eu/collection/open-so...
Everyone needs to stop competing with Office and just build a better office.
The truth is, no one needs to be compatible these days. Everyone is either:
- using the same software company-wide, whether it's LO, Google Docs or MS Office
- exporting to PDF when sharing docs with someone outside the organization
The real thing that LO is missing is server side hosting and easy syncing... Until they get that, it's just going to be something used by individuals and small orgs.
I still remember, at my university we had LO installed on lab computers, MS Office was a "requirement" for students to purchase, but most profs simply insisted we simply hand in assignments as Google Docs links because there was incompatibilities between .doc, .docx, the formats made by LO, etc... Google Docs were the only ones that could be shared and be 100% identical on every computer with the link.
If LO had a web hosted solution and provided easy to install server code for organizations, they'd dominate. But they don't...
LibreOffice has both of those, provided by Collabora (the main corporate developer of LibreOffice).
Hmmm... They probably need to advertise that more, or connect the two. Had no idea they were related. Also the Collabora website is cancer, it's not easy at all trying it...
I tried really hard to use Collabora 'web' solution. It is unusable.
For those who don't know: it renders the whole LibreOffice interface on server and passes it to you. Lots of issues with hotkeys on non-standard layouts (I use Dvorak), with language switching, with mouse behavior, with clipboard, ugh. Of course it is hungry for resources. No. This isn't the way.
Are you sure that's not just a VNC-based demo of the desktop version of Collabora Office?
I wouldn't be too surprised if their web version was like that (given how long ago the first version came out), except Nextcloud uses a Collabora-based system for its online editing, and I doubt they'd go with something like that.
There are online ODT editors (e.g. https://webodf.org/demo/ci/wodotexteditor-0.5.9/localeditor....), so even if Collabora's is bad, it wouldn't be too hard to make a better one.
Never attribute malice where incompetence is a sufficient explanation.
Microsoft isn't intentionally obfuscating the docx. Docx is a shit-show because hundreds upon hundreds of Microsoft business initiatives, executive pet projects, and ancient compatibility rules have all collided to make a giant pile of dung.
If LibreOffice is worried about what docx does to their productivity ... you should see the fucking engineers lamenting INSIDE Microsoft about what it does to their friggin productivity.
This horseshit isn't anyone's plan. This horseshit is an emergent phenomenon like a fucking termite hill in your back yard ... where no single termite is responsible or knoweldgeable, but all of them together made a pile that breaks your lawnmower.
I doubt this one is explicit "locking in", as much as it's reflecting the increasing internal complexity and lack of focus on product quality. I'd be willing to bet the internal teams dread working with the overly complex structures too.
At this point I doubt Microsoft could even execute on a "lock in" strategy anymore.
I know about the lack of tech-savvyness of most humans, but isn't Markdown and Pandoc—if you slam a GUI in front of it—covering the needs of 99% of users?
Granted, when you need formatting, like for a formal letter, you use a template someone made but this is not what most people use Word for.
And don't get me started on "people wouldn't understand how to put things in bold or italics"; they can barely use Word anyway. Might as well use something much simpler. Office "productivity" suites are over to me.
If you have such little regard for formatting that markdown suffices, you also don’t care about the difficult edge cases of docx.
* https://keenwrite.com/screenshots.html
* https://www.youtube.com/playlist?list=PLB-WIt1cZYLm1MMx2FBG9...
Here are some of my Markdown documents:
* https://impacts.to/downloads/lowres/impacts.pdf (99% pure)
* https://pdfhost.io/v/4FeAGGasj_SepiSolar_Highlevel_Software_...
* https://dave.autonoma.ca/blog/2020/04/28/typesetting-markdow...
A lot is possible with Markdown, especially with pandoc extensions.
(1) These look great - I saved the Impacts so I can keep reading it later. Can you share what the background/reason is for this document?
(2) Hoping you might be able to point me in the right direction. I often take notes with clients in Markdown, and then after the call, do some cleanup. At that point, I would love to export it as a slick-looking PDF, and also HTML which I could paste into Apple Mail. I've done a few experiments with Markdown --> Pandoc --> HTML, but I started to realize I would need to dig deeper into Pandoc, figure out its HTML templates, create CSS. And wasn't sure if I ought to be doing Markdown-->HTML-->PDF. My ideal "flow" would be something like...
pandoc 2025-07-18_Client1_Topic.md --style "Client1.css" ---output 2025-07-18_Client1_topic.pdf --output 2025-08.18_Client1_Topic.html
When digging around in Pandoc, I also got the impression I would need to learn Tex formatting. I haven't figured out a "stack" on this topic, so thought you might have ideas.
Currently, I am using Obsidian for taking notes. Then I use the "Copy as HTML" 3rd party plugin to copy my Markdown notes as basically themed HTML. I then open up MacOS TextEdit, on Rich Text mode, and paste that in. I might do a bit of text formatting here, clean out [ ] markdown links, etc. Next, I copy & paste from TextEdit to Apple Mail. I sometimes color the headings manually, and other basic formatting. Here I'll add the "Hello, " and "Sincerely" wrap around the notes, and send. I'll then check my iPhone Mail to make sure the mail looks OK on Dark Mode.
Seems like a huge gap in the "note taking" tooling, but I'm hoping that I'm just ignorant of a few solid tools that can handle this workflow: "Take notes on a call, send them nicely formatted to client in email / PDF"
Thanks for any ideas / suggestions you can share!
// JRO
cmark is an easy to build C project that reads markdown (commonmark, but close enough) that I also write in obsidian, and writes it as xml or html. You can probably automate some of that conversion process, e.g. with xslt, though I'll admit that's an acquired taste.
WordPad, Notepad, or TextEdit is already included on the computer and perfectly fine for most writing and printed communication.
I can confirm. When trying convert simple Word sentences and tables to e.g. Markdown/HTML from a Word XML you need a PhD in XML edge cases and nested garbage.
I wonder if this tool by MSFT is able to handle that:
https://github.com/microsoft/markitdown
I was amazed when I realised that Word docs were just zip files and you could poke around in the xml files embedded inside of them.
I almost implemented a working React -> Word document renderer back in 2017, but it didn't have support for creating the xml tags with : inside of them (which OOXML documents use).
Even though markitdown is a Microsoft project, it's just a thin wrapper around a bunch of 3rd party Python packages. For example, to go from docx to Markdown, it uses mammoth to convert docx to HTML[0], then uses markdownify to convert the HTML into Markdown[1].
[0]https://github.com/microsoft/markitdown/blob/da7bcea527ed04c... [1]https://github.com/microsoft/markitdown/blob/da7bcea527ed04c...
Technically, they're a bit more than just zip files (they're OPC containers [0]), but if you're hand editing the file content it doesn't really matter.
[0] Open Package Convention: https://en.wikipedia.org/wiki/Open_Packaging_Conventions
Well, it is not pretty to see how the sausage gets made, but extracting formatted text from docx is absolutely doable, no PhD involved. Source: I have done it as a little sidequest because it was useful to audit a set of word documents.
I loved OpenOffice.org back in the day, but in today's world with no modern web collaboration option it is dead. I'm aware of the attempts to make it run via browser, and no, they aren't really there, not even close.
What we need today is a web-first suite of apps that does everything Google Docs/Spreadsheets/Slides do, but uses OpenDocument family of standards as their native file format.
It isn't dead because it can still be used for many use cases. I write many documents in LibreOffice Writer and I regularly use LibreoOffice Calc for financial calculations. Also, LibreOffice Calc is a stupid name because it conflicts with the calculator in application launchers that search for substrings.
The last issue is trivially solved with a differently named alias?
I use Libreoffice at work to open and edit MS Office files and it works mostly fine, except for PowerPoint (missing fonts but also general madness). In general, I think the LibreOffice folks did a great job, esp. when it comes to MS file format support. And of course they need to, because I (and I guess most people) actually open more Ms files with LibreOffice than files in LOs native formats.
I'd really like to see an office suite which uses .md and .csv where possible.
Mostly I use LyX and pyspread which are close/open enough.
These are too primitive for an office suite
> I'd really like to see an office suite which uses .md
It is called pandoc and a text editor.
Csv is a portal to hell. Markdown is great.
After having worked for many companies, I think the complex file formats is likely due to employee turnover, pressure from management to just get shit done, and careless coding. Not ill intended anti-competitive practices. The locking in is likely an unintended bonus for Microsoft.
I didn’t realise we’d entered a time machine to 2007. I’ve worked extensively with OOXML and yes, the documentation is cryptic and often absent, but Microsoft will help you out if you contact them on their forums. I see Libre Office devs there all the time!
But the complexity is not some kind of conspiracy - it’s inherent - it comes from the fact that Office is ancient and very, very complex with a huge number of features. Many features are implemented in backwards compatible way on top of the old version of similar features and then the whole thing has been back ported from a bunch of C structure to XML which has the most woeful and underpowered schema language imaginable.
> it comes from the fact that Office is ancient and very, very complex with a huge number of features.
IIRC one of the many unfortunate decisions made by MS with OOXML (whether intentionally, or not, or both) is to codify a lot of display and formatting quirks directly in the schema with very little explanation or docs. Instead of making it s different namespace or layer.
So, to implement OOXML, you also needed to reverse engineer, say, behavior of Word97 etc.
Meanwhile, has a better modern featureful extensible rich text format been invented?
asciidoc and html?
Isn't it the same for PDF format spec?
The PDF spec is well designed to be both very expandable and easily readable.
It's absolutely horrible to try and edit.
That's because the structure of a PDF is essentially a bunch of media "streams". It's very easy to say "render a jpeg at this location on the page" but that's about it. It doesn't store, for example, the fact that you might need to wrap words around a page. Instead, it's "Here's a box with text in it".
The only thing that really could make PDF rendering hard is adobe put a whole bunch of garbage into the spec. For example, the full spec had the ability to run javascript and flash at one point (not sure if it does anymore).
JavaScript must still be there, because I think that's how form validation works? Don't recall Flash ever being there though.
It was acrobat reader, but not in the standard. For some time only the reader was needed and not the player.
No. There are hundreds of programs that easily read and create PDFs. OTOH, reading .docx is a pain. Far be it from me to defend Adobe, but PDF is nothing like MS Office formats.
Ooh that’s not fair. Many PDFs don’t conform to the spec and how Acrobat processes them is completely undocumented.
It was funny for a few years when Apple just released Preview, and it rendered many PDFs better than Adobe's own Acrobat
Yeah it's a great app, one of the big reasons I still use MacOS - but there's certain features which you still need Adobe Acrobat for, I know to fill in the tax forms for my country you do.
I mean the rendering part. AFAIK there are no that many renderers.
No, that spec is outright nice if you consider the PCs it was made for.
groklaw.net dived deep into this. AFAICR, MS never achieved round tripping between their own formats, let alone ODF. It was a lock in scheme where MS would delegate people to the ODF committees (don't remember the official name) to sabotage them from the inside.
Tangent: Is there a Firefox extension that blocks articles with slams, claps back, rips ?
Microsoft doesn't lock in people.
People lock in people.
You might not be wrong, but yelling at people for accepting conveniences that most don't even understand or care about, seems a bit pointless to me. Why even stop at document formats? Why not outrage over their choice of Windows itself?
What good do you think this does? I'm genuinely curious.
Well, according to StatsCounter, Linux now accounts for 5% of desktop users. :-)
ChromeOS another 2.7% and macOS around 24%.
edit: If I were to guess, Valve/Steam is solely responsible for at least 1 of those 5%.
> Why not outrage over their choice of Windows itself?
Because it doesn't necessarily affect anyone. Using Windows ipso facto doesn't mean you will send someone a file they can't read without a Microsoft program.
I have two Windows machines in my home; they have LibreOffice on them, as well as Firefox.
No, but I think the people most likely to be oblivious to all this, will be the ones using Microsoft programs that write in proprietary formats by default... and I would say they're not even wrong if that's what most other people use too.
People send me Outlook messages all the time that I can't read at all, and many Excel spreadsheets and Powerpoint presentations with features that aren't supported in anything on Linux. I literally have to use a Windows VM with proper Office to read this stuff. And not only work documents, but documents from my children's school/teachers as well.
[dead]