I use LibreOffice personally, and I'm thinking about moving the team to it before we seriously grow.
Has anyone tried this in their companies? Was it successful? I'm especially interested in companies of a similar size (or larger) than ours, and those where not everyone is super technical / a software engineer.
My big concerns are support and training, I think there are companies who offer this which would make it a lot easier.
(Edit: to be clear, our CFO will probably continue to use Excel, as will some of the team who build very complicated modelling sheets as that is what they're used to, but I'm thinking for the rest of us who mostly need decent spreadsheets, and good word-processing and presentation tools).
I don't believe your conclusion follows the premise -- I'm not planning my business around an exit strategy, and we're gunning for being a large, private company not just a quick money spinner.
As I've said elsewhere, it's also not about cost, but about open tools and an open ecosystem. I use LibreOffice myself (I run Linux, as do about a quarter of my engineering team) and find it fine (except for complex models and spreadsheets).
In your intro you mentioned a series A funding and so I imagined the typical start-up which goes maybe to 50-500 employees before exit (or bust). If you are gunning to be a next RedHat or strongly driven by a Open Source strategy, I get why you would strategically invest in being a pure Open Source shop. If Linux drives your business then it is fine to go this way. Otherwise: The office stack is not a core part of your business so go with the thing that causes the least amount of friction in your company for a modest price.
Use the thing everyone knows.
Why do I want to look at LO?
- GSuite: it's just terrible, the entire suite misses basic functionality (custom styles, anyone?) and it only works online for me, and there are no files I can access. When Google went down for a day no-one could access a single document or presentation, which is just untenable.
- MS Office: better than GSuite, but the latter has better collaboration tools IMO. I like Office just fine, but it's not OpenSource, and I'm on Linux which means there are no offline apps I can use.
- LibreOffice: it's open-source, with an open document format. If we don't start using (and paying for) open-source projects as business they'll never become sufficiently competitive, and I prefer an open ecosystem to a closed one.
Style management stands as a particular example. Now you must wade through dozens of unwanted canned styles to find the ones you've set up... if I remember correctly.
For the style mess and other UI regressions that I haven't taken the time to catalog, I switched to LibreOffice. Granted, I don't exchange Office documents with coworkers much. We use (also crappy) Confluence for most documentation; and if I just want to send a document to someone, I export to PDF.
I used to love Word. I used WordBasic to automate not only technical-documentation production, but also to parse and rewrite thousands of SQL routines on a DEC VAX system and save my company weeks of engineering time.
I hate Word now because Microsoft ruined its UI. Oh, and I detest the continual hounding to log into your "Microsoft account." Nobody wants your goddamned account, Microsoft. We all have PLENTY of accounts by now.
"I think your tabstops are inconsistent."
Granted, I don't do complex spreadsheet formatting.
Are there open-source implementation of Docs, even poor ones?
Yes: https://cryptpad.fr/
- Lost modifications, a lot of times ("something went wrong, please reload" and then the last 15 minutes of changes are just GONE)
- Very frequent "synchronizing... please wait" screens that block you from doing anything for a few seconds
- Undo/redo not possible in "live editing mode"
- In "non-live editing mode", if you don't hit save manually and the TCP connection resets, you will lose all your changes
- Pivot tables do very weird things with formatting
And that's the experience of a solo user, there wasn't anyone joined in on the document at the time. I can only imagine how it would have been if were multiple persons working on that spreadsheet.
I do have a flaky Internet connection, but Google Docs handle it fine, and emacs-crdt works very well too.
I'm now refactoring the whole spreadsheet into a bunch of .csv files stored in Git, and a Python script that outputs a pivot table, because I've lost more types due to CryptPad bugs than the time it would take me to write that script.
I LAUGH at silly FUD like "you can't run a business without MSOffice", "You can't read documents ", "format compatibility problems!!", because that all is just FUD.
If I ever (rarely, maybe 4 times total) receive a price file or spec that is unreadable by LibreOffice, I just ask to resend it to me in some open format cause I lack the software to open their proprietary file. And I did received it in a PDF or .odt every single time I asked.
That said, I liked the software more in their OOo days. When it became LibreOffice it became somehow more buggy and ugly. OOo with the right icon pack was very neat and had a superior interface to MSOffice. LibreOffice icons got uglified and are no longer pleasant to look at, and combined all looks... Worse.
Their online efforts are also disastrously bad. LibreOffice Online is based on a flawed idea to run LO on a server and deliver rendered images to a user. It works super shitty bad. They should have just made a web-first editor for OpenDocument file format, with concurrent editing. (Maybe when I'm finally done with the XMPP I'll look for funding to make just this. ... )
Not really a FUD if you're dealing with Excel spreadsheets from other companies, especially those that uses integrations with Excel. However PowerPoint/Present compat is a hit-or-miss bother but not world-shuttering like with Excel though, and Word compatibility is actually okay except when dealing with weirder documents that is also fragile with different Word versions.
Regarding the presentation format, the One True Format is PDF.
Matter of taste - I quite like the current interface. I've just fired up Writer and it actually reminds me of Word 2 with a much higher resolution and a lot of polish. There's just enough functionality exposed via buttons and no more. There are menus with loads of functions and they are sensibly laid out.
LO is being developed at a remarkable pace. The codebase they forked from Apache that came from SO was pretty wank and had some fairly odd things going on. It's a massive job and I really like their direction.
The end result has just worked for me for at least five years now with no friction.
I still have a pretty nightmareish template of styles that started off life in Word 9x and still going strong. It was for ISO9001 docs and includes three layers of table styles alone. Our ISO9000/27000 systems are all on Mediawiki these days but that template still works fine.
There are some jolly exciting bugs in all software but I suspect you are not an actual LO user.
Not worth it imo.
https://www.zdnet.com/article/linux-not-windows-why-munich-i...
For majority of users the features office provides is a waste of their time.
The usage experience on MSOffice is unpleasant enough to motivate seeking an alternative.
Folks in tech tend to be more worried about UI and UX and frankly tend to be nit picky when something doesn’t look or work like they would have designed it.
The issues I face with my company are usually accountants/bookkeepers are used to using Excel. LibreOffice usually can handle anything they are wanting to do. It just means a little extra time googling/playing around with it to work the same way. I try to keep everything on Google Sheets however.
That said, if we were scaling post Series A (we are not), I'd look to see whether there are teams that need advanced modeling or feel comfortable with Excel. Everyone whose job is not spreadsheet wrangling can go with LibreOffice and Google Sheets.
Have you done a cost analysis of what the sum of that "little extra time googling" adds up to? If it's >$10 a month, then youre company is flat-out worse-off. Especially when LibreOffice / Google Sheets just doesn't even come close to the ease of use or power that Excel provides.
It's like giving a brick to a carpenter that has used a hammer for many years. Sure, the brick is okay at smashing nails in, but it flat out can't pull nails out, it creates ugly work, has no "user experience guard-rails", and is less efficient.
I don't really think there's anything I know of (that a normal small company would use especially) in which LibreOffice will not be as good or better than Google Suite, so if that's what your current office suite is then the transition should not be too painful. Googling for the answer works for most questions.
The office suite is better, the email client configurability is significantly better (from what they claim, I'm a Thunderbird on Linux person), and you get single sign on too.
My firm uses our own mailserver (with DKIM and SPF), LibreOffice for doing our cost comparisons for clients and other document generation needs, and has been pretty happy with that. We are split across Windows, Linux and MacOS, the sole tether for about half of our Windows users is a dependence on iTunes backup (since apparently they don't trust iCloud, ugh...) for their iPhones, preventing a migration to Linux.
I'm gonna have to disagree on this one. Outlook 365 is ass-tastic. It doesn't even have feature parity with the web app, and is full of weird bugs and limitations. Just off the top of my head:
- Moving old messages between folders in the native app requires downloading the full message then uploading it again
- Calendar subscriptions sync from web client to native app, but not the other way around
- Calendar colors on the native app are limited to a preset pallete of 8, while the web app allows custom colors
- The new mail notification icon sticks on if another client reads the new message
- Cannot sync certain types of rules with the exchange server
- Warns you about "dangerous attachments" if you open an attachment that _you_ just added to an unsent email
- No way to block specific addresses / domains besides a local-only(!) rule to always put them in junk
I have worked at a company that started with Googledocs and ended with Markdown (paper / design docs) and reveal.js for presentations. And that is something I can take recommend. Why? Because it is simple text that can be stored in a repository and you gain the ability to compare versions. It is easy to change and you can convert it to almost anything. The most important target is obviously pdf to communicate with customers. Nobody wants odt, docx or other strange formats that require more than a browser.
Convincing the technical people was not necessary or done quick. Non coder like the CEO did like the suggestion least, but it was a revolution from the base and people who code where mostly creating content or talking at conferences.
It only makes sense of you can convince people.
There was thrift in my hometown, that was using excel to decide on the max credit for individuals. The document was copied and forked and the outcome was more random than anything. At least they could be convinced to introduce a database and a well written and maintained application, that does reliable work and provides the same output when given the same input.
She recently got a job and has to use Office but her LibreOffice experience hasn't set her back.
Don't let the naysayers stop you.
(Random additional comment: One thing I especially like about LibreOffice Calc over Excel is its process for importing CSV files).
For text, Markdown with inline semantic HTML if necessary and CSS stylesheets if needed is better, since it properly separates style and presentation unlike word processors.
For modelling, Jupyter-like notebooks are much better than spreadsheets since they properly label objects, don't make everything a 2D grid, and use a proper programming language.
For data, databases are better than spreadsheets.
Presentations are just documents, so there is no reason to use separate software.
For drawing you can use Inkscape and Gimp.
For collaboration you can use normal distributed version control like Git (with solutions like self-hosted Gitlab).
Except maybe Gimp? Most people in offices don’t do image editing…
Any suite only exists for it's output. If you use a generic format then you can convert to specialised ones as required. If you save in a proprietary format, be prepared for vendor lockin.
Say HTML5 for docs, some form of CSV/YAML for spreadsheets etc you get the idea.
Keep your data, then your suite decisions flow from that.
But consciously choose the file format first. Then the suite. The file format is where your info will be stored.
This isn't about cost, but as I said elsewhere:
> I have no aversion to paying (we pay Google, for example) but I like open source, would like to contribute (financially) to LibreOffice or whoever is relevant if we do go down that path, and think that the only way this stuff gets mainstreamed is if companies, y'know, mainstream it.
Also, what makes you think LibreOffice doesn't give users smooth experience?
Are you donating towards the cost of maintenance and development?
If not, what changes could they make to convince you to do that?
If you find an awkward corner case in any piece of software then you will find support rather lacking from the vendor. I use LO myself: I'm a business owner and can use what I damn well like! I don't force anyone else to use LO but we are a MS tin partner or something so get our tricks err drugs err software a bit cheaper.
Training: I was a trainer back in the day and it's all about concepts and not the implementation. I've used Word from v1 to date and 2.0 was a sweet spot in terms of functionality and performance (2 floppies worth of code but could do quite a lot of decent WP and rudimentary DTP stuff eg hanging indents, decimal tabstops and text/image frames with text flow. Modern LO Writer has a turtle implementation in it, coz reasons. Nowadays I struggle a bit to find things in the ribbons. Super Calc, Lotus 1-2-3, Excel and LO Calc all do the same thing in the end to a greater or lesser extent. I wrote a Finite Capacity Planner (yes: Finite) in Excel for a factory around 1995. I'd have to rewrite my functions in Python or whatever but it would run in LO quite happily.
Another anecdote: My wife is not IT minded in any way as such and has worked as a classroom assistant in schools catering for special needs and adult support services etc for several decades. She has had various IT offerings foisted on her from Acorn Axxx machines to PCs running Win 9x etc. At home she used to have Windows on her laptop but I eventually had enough at Win 7 when yet another graphics "driver" update needed a good six hours on the operating table to fix. She now runs Arch (actually!) and doesn't care. It just has to work. She also uses LO, which seems to work OK and isn't the oddest office suite she's been forced to use.
Don't overthink it. LO Writer and Calc are extremely capable apps and Base looks like a pretty good effort too when I've used it and you get all the rest of it but I tend to avoid Presentation type apps (such as "Poor Point")
If you are capable of running a start up then LO will work fine. If some of your staff are incapable of any flexibility - I'd start to reconsider my choice of staff. A startup is all about being flexible and seeing opportunity and so on. If any modern IT user is "incapable" of getting on with LO then I'd question their ability to work in a startup environment.
The issue isnt whether LO is good (it is) but whether the LO team can compete with a trillion dollar corp that considers 365 as a central pillar of their business, invests in it accordingly, and whose changes to the software become the defacto standard overnight.
Microsoft consistently shoots themselves and their customers in the legs in desperate efforts to ape the young cool kid hipster companies.
I think the real question is: how long can a company trust a sinking ship, that, while it still has plenty of money, is bent on destroying everything they built so far?
Abandon ship! The captains have gone mad.
But really, probably none of that matters. If you think Google Workplace is too clunky, you’ll never be happy with LibreOffice. Just get O365 and get back to work. It’s not worth saving ~$5k annually.
For the last 4 years, I’ve been CIO and VP Eng of a now-series B company. We’re at about 70 FTE. We use Google Workspace and O365. We tried to move everyone to Google, but we couldn’t do it. As you noted, there’s just too much missing from it.
Sadly O365 seems no-go on Linux :(
The only slightly related issue has been the one client who continues to send email attachments from Outlook as winmail.dat.
Example, at our office we have a standard way of sharing info (Slack), and store shared information in Notion. But some folks use Google Docs/Sheet, some prefer Excel, or Markdown files for their on work uses
I like Google a little bit for its ability to remote code the spreadsheet and the sharing feature, but now in a large financial institution, Excel works perfectly fine.
I suppose the Word and PowerPoint alternatives are not as bad in LibreOffice, so if that's your main tools, go for it.
When she started to work for she asked me if we have LibreOffice on our computers because she had to edit some documents she took from work. I was pleasantly surprised to learn that they relied on LibreOffice for document management. Nowadays LibreOffice is installed on all our home computers alongside Microsoft Office as well. She can switch to whatever she finds easier and I taught her to always convert her documents to .RTF to be sure she has no problem switching between home and work.
I have a similar use case (plus add in cafes where I'd rather not be on their wifi or my own mobile data). It amazes me how many company IT people I've spoken with over the years who just don't comprehend some people work without being connected to the internet 100% of the time.
Dropbox remains my solution there. I'm also MS Office - too much interaction of documents with my clients to try anything else.
-the report generated by our laser alignment machine is in a 1993 ms write format. o365 says its no bueno.
-the diagnostic tool we use on some truck engines for valve calibration, advance and timing spits out an interactive doc in some esoteric ibm format I think? we edit the output in libre, feed it back into the tool, and it makes our adjustments.
-our shop floor uses libre because we don't have enough internet to run o365 without huge delays. we switched four years ago.
-our ancient jit parts order system will gladly handle lotus dif files, so libre does too
I’ve never seen a single person react to it’s apps as being unfamiliar or different, especially considering the variety of UI shifts M$ Office has displayed since Office 2k. If you were referring to some specific API and integration with Sharepojnt or something I might understand your trepidation, but as your question stands, I have to say that I can’t think of a specific downside to it.
I suppose that deciding upon an output format that you will send out to clients or suppliers will be your main concern, as you will essentially choosing which version of M$ Office you are pretending to be. I find that the older pre-xml format is more universally supported but YMMV
But LO Calc is bad, or at least was when I was trying it last time. It has been creating broken tables especially where filters / conditional formatting is involved and was endlessly crashing on attempt to create graph from 2 channel scope (cca 2*2000 X-Y values). OTOH in Excel I can have table with more than 30k lines with filters, sorting calculations and everything works swiftly. No, I am not going to fool around with a database.
In Word vs Writer scenario is not anything really appealing to me as I don't use them too much. So I am neutral on those.
Recently, frustrated again with some excel shenanigans, I looked around and I discovered softmaker office. I immediately got a license and experimented a little. It seems very compatible with ms office. It still costs money, but less then ms. So in the near future I might try that route, along with thunderbird to replace outlook.
I feel like LibreOffice/OpenOffice file compatibility with MS Office peaked in the early 2000's. The last time I tried to load a Word doc in LibreOffice a few years ago, it looked very janky. About the same as it did in 2004.
Sadly, I still receive 2 or 3 Office attachments a week. So MS Office compatibility is still very relevant.
To be clear, as your edit says, your CFO and other modellers are probably going to have to use Excel. There are just things that work in Excel and nowhere else.
Writer is a problem in that it is a straight rip-off of Word, which supports the same file formats, but which is not fully compatible with Word.
Word and Writer handle styles differently: glyphs change, despite availability of the same fonts; alignment changes; tab stops are all over the place.
They handle images and floating things differently, sometimes placing things on different pages or behind other objects or text. Word online makes this even worse, because it doesn't handle image captions properly, and converts the image and the caption to a second image.
I get all sorts of weird problems with Writer and Word - no particular blame attached - and unfortunately the answer is always the same: do your work in native Word on Windows, unless all your work is going to be in LibreOffice.
I think that the basic premises behind modern word processing are just wrong, and it's because Word is so ubiquitous that everything else has to follow it. I think there is a space for a new style of WP which takes nothing from Word, or Writer, and which lies somewhere between that point and professional typesetting wares like LaTeX or InDesign; something which does less more professionally.
But it's just a dream, and it doesn't really take into account Excel, which is Microsoft's killer app and probably the keystone of the company's entire business.
Then of course there's the whole parasitic, prying surveillance and snooping in the name of "helpful" user experience enhancement, telemetry and other bullshit corporate-speak for shamelessly sucking up your data. Another awful-smelling infection spreading among all the big corporate software players and making MS Office in its latest versions much more like the abomination that is GSuite (or anything by Google really). Do you really want most of the documentation being created by your business to be subject to perpetual viewing by these companies? I'd hope not.
Libre Office isn't perfect, but at least it has much less of all the above while still keeping most of what makes an office suite simply work across the board.
Hiring someone to translate those "complicated" sheets in to a combination of python and basic CLI tools like grep to work with CSVs might be worth it in the long term, but for word processing and presentations, if you can export to PDF rather than be passing things in and out of the organization, OpenOffice should work well.
Your issue will be that ultimately, if you're not exporting to PDF or ODT but trying to export to Microsoft's proprietary format you'll always end up chasing some obscure bug on a deadline that will make the entire endeavor not worth whatever costs were present up until that point.
And the CFO uses Google Workspace as well.
I always use the same software that the C-suite uses for those files we transfer between us.
I used Abiword and a lot of Gnumeric in the past. Now everything written in emacs, and then directly glued to postgresql, R or Perl. I hadn't opened an excel since... one decade maybe.
I agree that for finance employees, Excel is way better than anything else available.
Offline - works fantastic tbh. Tons of experience on long haul flights