A previous company adopted OpenOffice for all internal uses, and our employees quickly came to love it. We had a customer once who sent us particularly tricky documents, and eventually we gave up and started using Word to manage their stuff. They still complained at broken formatting after we'd edited their docs and mailed them back, though. The conversation went like:
Us: That's really strange. We were using a different word processor, but you're still seeing the same problems even after we went back to Word.
Them: Um, what word processor would that be?
Us: You probably haven't heard of it, but OpenOffice.
Them: Us too! Can we just send you ODT files directly?
The workflow on both ends had been to work natively in Writer, then export to .doc for passing files around. All problems immediately went away once we dropped the middle step.
I outsource some of my translation work on oDesk and ask that all of the work be done with OpenOffice. Many already have it but some need to install it. If everyone works like this, we'll have a lot more people using Open Office.
"It hasn't been feasible to uninstall Word, or abandon it entirely. Of course there are times when we need to collaborate with Word users, and perhaps a quarter of the time we work on that platform."
I've used the StarOffice/OpenOffice/LibreOffice line for a decade. The last Microsoft Office product I purchased was Word 97, and I ditched that when I retired the Windows 2000 machine.
That said, LibreOffice is still too buggy for a 10-year-old product. The article says "Writer has a couple of instabilities under Windows, such as locking up if a new document isn't saved within a few minutes." That's pathetic.
LibreOffice .doc import/export almost works now. You can usually import a Microsoft Word document, although the margins are often wrong.
LibreOffice attempts to replicate the MS Word "experience", which is good if that's what you looking for, but which I find atrocious.
Do you have recommendations for a word processor which
- Focuses on writing text documents, doesn't attempt to do everything
- Is visual, because sometimes it's painful to go through cycle of compilation with LaTeX
- Has a minimalistic UI that focuses on semantics; which lets you mark things as sections, titles, emphasis, etc. Fonts, colors, are handled separately within a template editor
- Ideally, allows for real-time collaborative editing
I've switched to using Markdown for almost everything I write. Take your favorite text editor, use a nice preview app (I like Marked 2 for OS X) for tweaking, and enjoy plaintext, grep-able, Git-able, docs that are trivially readable even without a special viewer.
That's obviously not an option for everything, but I've used it for everything from internal documentation to slides to blog posts.
> sometimes it's painful to go through cycle of compilation with LaTeX
I've found that successful use of LaTeX requires you to avoid compiling on a regular basis. When I write something in LaTeX, I chunk the time I spend on formatting/compiling, so that it doesn't interfere with writing. I also find it helpful to construct multi-file templates that enable modularity and reuse.
The only time I really use a GUI is for formatting equations, since being able to see the change right away is very helpful.
Have you tried LyX (http://www.lyx.org)? It's a WYSIWYM (your first point) graphical editor for LaTeX and as far as I remember shows you a pretty good preview of your document as you're typing (your second point).
Has anyone heard of professional use of LaTeX outside of the scientific fields?
I've seen a lot of good examples of the power of LaTeX but all seems ot have been made by scientists. And I tried to convince my girlfriend who is in the edition/layout field (not sure of the correct name). she already knew LaTeX (at my surprise) and she thought the examples were unimpressives. She doesn't anybody in her field using it professionally.
So I was wondering if her experience was specific or it was really unused outsite science. Any feedback on this one?
The answer is a resounding 'no'. Typesetting is something that is nearly completely done by InDesign at large shops — the advantages of LaTeX just aren't appreciable when laying out photo-heavy books that use layouts custom to each book. 99% of the things InDesign can do are definitely replicable in LaTeX, but it'd be a ton of work and would require one hell of a knowledge of LaTeX libraries. Contrast this with the fact that a competent high schooler with some knowledge of design can pump out great looking layouts with InDesign. LaTeX is great for personal documents and publications where layout has been wholly standardized, but those situations are quite rare for publishing firms.
LaTeX is pretty dead outside of science, and it makes sense that this is the state of things.
Using LaTeX effectively requires the same ways of thinking needed in order to write code. For people who aren't used to doing this, adjusting from Word to LaTeX is not going to be easy.
In the professional publishing fields, editors have overcome the shortcomings of Word in a completely different way - using advanced GUI tools that have hundreds of options and settings that you have to learn, but that let you have instant feedback on your decisions. Admittedly, programs such as InDesign can actually make prettier documents than LaTeX, but the comparative amount of effort you need to invest makes it rather a bad tradeoff for someone who already knows LaTeX.
LaTeX is good at making consistently beautiful documents fast, while professional publishing tools let you invest lots of effort into getting something even more beautiful, or downright ugly. For the scientist, LaTeX wins easily, but for someone whose primary job is to make things as pretty as they can be, it is not so.
There's a lot of unprofessional behavior and software practices in the "professional" publishing field. Many authors refuse to write in anything but MS Word, and companies rely on Track Changes for version control of manuscripts.
I'm definitely not an everyday user of Track Changes, but isn't that the very use case for it, version control inside a manuscript?
The biggest caveat I've seen to its use are ballooning file sizes (meh), and (more importantly) leaking information. But I also thought recent versions of Word made it more prominent to discard tracking information when doing a Save/As.
I use Windows x64 and for the past one and a half year have kept it MS-Office clean. However my colleagues usually argue with me about this. Even If I try to explain them the advantages, they keep insist on using MS-Office even for scientific papers where I find Lyx superior. At work I installed it since I do not pay for it. At home, never even If I am on windows.
My only complaint is the lack of recent builds of Abiword/Gnumeric.
Abiword is on 3.0.0. in mingw64 in OBS.
Gnumeric is even older.
These problems have to be fixed.
Does anyone know where I can find recent builds for windows 7? x86/x64 makes no difference to me.
What about the following combination: First, set up a Windows based X implementation (Cygwin X, or Xming for example). Then run a stateless Linux install in a VM (maybe based off a live CD), and then set up icons on your Windows desktop to remotely launch the given apps on the Linux VM, displaying them on your Windows desktop with your local X server.
The main problems I see with this would be automating the Linux VM startup, and configuring the local icons (they would have to launch a script that ssh's into the Linux VM after checking if a local X server and the Linux VM is running, forwarding X connections).
I am convinced the majority of the software in the world is really written in Excel, which was ripped off Lotus 123. LibreOffice does not really compete.
I made a switch to LibreOffice when my work machine switched to Ubuntu. As the post points out, switching for everything but Excel is feasible for most users. There is a bit of a learning curve, but some things actually seem to make more sense once you dive in.
As for Excel, as tumultuous a relationship as I have had with that software over my working career, I never really appreciated its features and stability until I tried to use the alternatives. That said, I haven't tried 4.4 but look forward to giving it another shot.
We simply don't apologise for jumping ship from MS Word. For us, the productivity gains have been irresistible.
That's awesome. I wish I could. I'm a technical writing consultant who does grant writing for nonprofit and public agencies (www.seliger.com if you're curious), and everything we get from clients and public agencies is a Word or Excel file. Everything we send out has to be in the same formats.
We're in a situation similar to the graphic designers and photographers and so on who say that everything they take in and send out has to be Photoshop / InDesign / etc. compatible. The industry has standardized around these programs and formats and perfect compatibility is mandatory.
LibreOffice is great. I've mentioned it in class (I adjunct on the side), but I can't use it.
There WAS a window, but I think it's closed already though. Google Docs and a host of Office-compatible Android/iOS apps made a go at it, and they made headway for a while. But Microsoft went ahead and made web and Android/iOS versions of Office that are at least as good as Google Docs and the various Office for Android apps I've seen, with far better Office compatibility than anyone else can offer (since they're running the same code as desktop Office under the hood).
Meanwhile LibreOffice is fighting the last war, with a thick desktop client using some custom cross-platform GUI toolkit that looks native on exactly nothing, no web app version and no real effort on mobile[1]. And it's fighting the last war badly at that -- not even people who write articles about switching to LibreOffice can actually switch to LibreOffice because they keep Word around for compatibility with everyone else in the world. Calc versus Excel is in even worse shape than Writer versus Word.
Does it finally do footnotes in a decent fashion?
Literally the only reason I kept a portable version of Word 2003 around during college was to do the final formatting and citations for my papers. I nearly broke down and learned LaTex instead, but that curve was just a little too steep.
That is the primary reason I keep office around for as well. No alternative offers decent footnotes / endnotes / reference without going to latex (that I have found). For something as simple as a two page paper for an undergrad class, I do not want have to use latex just to get footnotes done correctly. I would be interested to know if the latest version actually does any better.
What difficulties are you having with footnotes? I just finished working on a 200 and something page document in LibreOffice 4.3, with multiple footnotes on most pages (and mildly complicated things like re-starting the numbering for each chapter), and working with footnotes seemed pretty straightforward. I'd be interested to find out how your (and the parent poster's) experience differed from mine.
Sadly, it still has issues with importing footnotes/endnotes.[1] It was one of the main reasons cited by our marketing team for resisting LibreOffice and going with MS Office 2010 instead.
LaTeX is totally worth it. I started off by using the LaTeX plugin in LibreOffice to do formulas. Eventually I realized that I could just do the whole paper in LaTeX. It isn't too hard and there is a stackexchange site for LaTeX that is very helpful as well.
What I miss, probably more as a PowerPoint competitor, is a Visual Studio/WPF-style editor with WYSIWYG and mark-up side by side.
I don't even want to think about the amount of time I spent trying to select some text I want to edit, buried under other transparent layers or empty, invisible boxes. It would make it also very easy to introduce dynamic references in the text of a textbox in powerpoint (something currently impossible!).
It wouldn't hurt the basic experience for normal users and would make PowerPoint infinitely more powerful for advanced users.
Word is less visual but this side by side approach would also be very useful. Particularly since a lot of the visual formatting is "invisible" in a WYSIWYG editor.
I'm not in a position to ditch Word, but I am happy to hear that there is a more robust alternative for numbered paragraphs with multiple sub-levels. Word appears to get confused when paragraphs change sub-levels.
LibreOffice crashes way too often for me to consider it a remotely acceptable product. Frequently it will crash within 10 seconds of opening a document (on Ubuntu).
If this is happening with the distro LO, then there's something seriously wrong with your PC or your installation. This does not happen for almost anyone else - whatever compatibility complaints people may have, LO is ridiculously stable. (The Coverity cleanup was seriously worth it.)
If you can reproduce this with a given document on a different PC, you have a reportable bug.
Most people can get by with Abiword and gnumeric. I often hear people say that there is some feature of MS Office that they need and that isn't provided by Abiword + gnumeric / OO / LO.
Is there a list of these features anywhere? And are the competitors to MS Office working on them? Or are they focussed on different things? EG a spreadsheet that focuses on correctness and auditing could probably be useful.
If you're using a Sharepoint server to manage documents across an organization, MS Office has Check-In & Check-Out capabilities integrated directly into MS Word. You're logged into the server through MS Word and you can manage visibility of your check-ins (ie whether it's a minor edit only visible to your direct collaborators, or a major edit visible to all people with access rights to the document).
If something like that is key to an organization's workflow, competitors will probably be locked out even if they do add support, because it's like using "non-genuine parts". It will probably work 99% of the time, but it's probably less reliable than the general part, and hard to get support for the cases where it does cause problems. Reliability & consistency is often more important than cheaper upfront cost.
(I say this as a long time NeoOffice fan & paid supporter of the project.)
I've noticed that none of the FOSS spreadsheets do as well with some things as Excel does.
For example if I put 1, 2, 4, 8 and I highlight that and drag it to continue the series I get different results.
Excel gets it right and provides 16, 32, 64, etc
LibreOffice provides 2, 3, 5, 9, etc
OpenOffice doesn't do any better.
I haven't tried gnumeric in the last 10 years so I can't remember what it does but I'm not optimistic.
There are a lot of these kinds of things where Excel really has polish and the others don't and probably never will because the code to make it work right is excruciatingly boring to write.
This is a pretty obscure YMMV, you must admit ... full quirk-compatibility is an iterative process.
What it'll take for LO Calc to compete with MSO is someone wanting to knock Excel off its perch and spend money on the drudge work. LO finally has decent data structures in Calc and the GPU calculation, now there's all the drudge to do to literally beat the best spreadsheet available. If you got the sort of serious spreadsheet that takes Excel 24 hours to run, and Calc ran it in 20 hours, people would pay any amount of money for such a thing.
(I think Excel is the hook, because it's the one bit of MSO that the people who sign the cheques use personally. Maybe Powerpoint too.)
As a word processor, Writer is every bit as good as MS Word, though of course slightly different. DOCX compatibility is improving noticeably every release; I haven't had any problems myself in several years.
Compatibility quirks are important to someone when they bite them, but look at Google Docs - it's hugely popular (basically for non-sucky email) even though its word processor and spreadsheet are terrible. We use GApps in the day job and I keep LO to hand for when I need something not terrible. Or that offers non-terrible DOCX rendering.
So I think what LO needs is something killer. Not sure what though.
Yeah, again it's a YMMV. GDocs' .docx rendering is dismally awful compared to LO, and using LO to talk to .docx is my main compatibility point.
What I really want is LO to read/write to Google Drive transparently, but that's still an experimental feature (though it works well in TDF builds, but only for password authentication, not for two-factor). Again, it awaits someone spending money on it.
We use them; I haven't had MS Office installed since... I'm not sure, but I think it's been more than a decade now. I used OpenOffice for ages, then LibreOffice.
For work we MS Office or LO for MS docs from elsewhere. LO is much better at correctly reading in MS-generated docs than Google Docs, and of course MS Office for people who spend most of their time dealing with documents from outside (not me, thankfully).
We use Google Docs for all of our internally-generated stuff, though -- the concurrent editing in particular is quite useful. We sometimes collaborate on docs, and each person can add their section at the same time. You can review someone else's work on a spreadsheet while they're doing it.
I recently went through an important (and politically complex) meeting where the CEO and I shared a Google doc for taking notes. It was awfully convenient; I could provide him constant feedback on technical issues but let him decide what to actually say, and how to say it.
I'm less happy with how Google Drive is integrated with Google Docs (not well), and I'd like to see concurrent editing in Sites, but maybe someday...
"Gets it right" is incredibly subjective, you cannot really expect a spreadsheet to run some number series analysis to see if there is some logic behind numbers except for some trivial cases (this one might be one, but that's debatable).
gnumeric does simply replicate the series, you get 1, 2, 4, 8, 1, 2, 4, 8, 1, etc. It does offer a tool to fill series though.
Although I think of the powers of two as the natural completion, OEIS finds about 1600 sequences that start 1, 2, 4, 8, and are of some mathematical or cultural interest.
It is subjective, but one of the things that normal users have come to like about Excel is that it does things that they would expect... Such as the number series incrementing which msandford described.
Maybe I'm remembering this wrong, but I seem to remember Excel doing some intuitive series extensions and both LibreOffice and OpenOffice not doing things that I would have found as intuitive.
They're probably not unreasonable fall-back strategies if something isn't strictly linear, but just because I understand why the programmer might have done that doesn't mean it makes sense in general.
IMHO it all boils down to habit. I switched from Excel to LO Calc and am quite happy, too, though it does feel a bit awkward and sometimes it just gets in the way. I managed writing a macro though, and it was surprisingly easier than I thought.
No—it boils down to network effects and coordination. As I wrote here: https://news.ycombinator.com/item?id=9069761 , in my business we use MS Office, our clients do, and funding agencies do. There is no easy or simple way to move everyone from MS Office to an alternative simultaneously. So it doesn't happen.
People who never or very rarely need to do document interchange may be okay with Google Docs or LibreOffice, but people who interchange complex documents on a regular basis must use MS Office.
Us: That's really strange. We were using a different word processor, but you're still seeing the same problems even after we went back to Word.
Them: Um, what word processor would that be?
Us: You probably haven't heard of it, but OpenOffice.
Them: Us too! Can we just send you ODT files directly?
The workflow on both ends had been to work natively in Writer, then export to .doc for passing files around. All problems immediately went away once we dropped the middle step.