LibreOffice 6.3 hits beta, with built-in redaction tool for sharing those █████ documents The Document Foundation has released the first beta of LibreOffice 6.3, with new features including a redaction tool and a Fourier Analysis spreadsheet function. LibreOffice was forked from OpenOffice in 2010, which is when The Document Foundation was formed, including a team of OpenOffice contributors. OpenOffice has also …
Actually I use OO's calc tool to manipulate enormous machine generated .csv files that cause Excel (any variant) to segfault and go down in flames. Hundreds of meg in a go? No problem, even on modest hardware.
For me, where OO really shines is the equation editor. Unlike MS Word you can compose complex equations without having to constantly bat your hand between keyboard and mouse. It's a beautiful design. Not as powerful as LaTeK, mind, but good enough for government work.
Where OO suffers is when you convert to and from the Microsoft file formats to the open formats; I waste far too much time re-doing PowerPoint graphics foo, redoing margins in Word, and so forth. Suspect it's a combination of differing font configurations between a Linux and Windows box, combined with non-standard or undocumented cruft in the MS file formats.
Hundreds of meg in a go?
You write that like you consider sub GB files to be large these days.
Excel handles hundreds of meg .csv no problem in my experience.
I do use OO and it is the equation editor that is the attraction too. However, everything still have to be loaded into MS Word, formatting checked and then saved in a Word format before I can send it on to anyone else.
Excel 2016 does contain significant performance improvements (fully 64-bit, multihreaded calcs) over previous versions, but use whatever works best for you. Not sure what you're doing with the CSVs subsequently but many people I know just use Pandas for the heavy-lifting and save to Excel for the manager-friendly reporting.
I concur, OpenOffice/LibreOffice circa 2014 UI and document rendering felt horribly slow on Windows (one could _see_ elements being rendered), contemporary MS Office felt so much faster in comparison on the same machine. Together with the extravagant-lloking UI, OO/LO was simply frustrating to use.
I checked latest stable OO 4.1.6 and LO 6.2.4 on the same machine: while still not instant, the rendering is much faster now indeed and does not feel frustrating. The LO 6.2.4 default UI looks so much better too.
I think the importing to any kind of digital format requires, of course, that the blacked out text really is removed and not merely concealed, and you also want to keep the flow of the document the same (not all letters are equally spaced). But this should indeed be possible when exporting to PDF using a graphic filter and LaTex presumably also has some filters that let you substituion.
███████ think ██████ one of the ████ creative ████ opening █████████ I've ████ https://www.youtube.com/watch?v=NBQJjqnG1iI
Bunch of overpaid and head-slappingly-under-educated-in technical-matters lawyers once sent me a bunch of redacted word documents, regarding technical security compliance in a Mergers and Acquisition situation. They had just highlighted the redacted text in black.
Select all.....
Reminds me of the "censored songs" round they used to have on ISIHAC (non-UK readers might as well give up now as you probably won't understand this!) where the teams were given songs to sing but had to use a buzzer to censor any words that might "frighten the horses" ... this reached a high point when Tim and Graeme were asked to sing a well known welsh song/hymn .... which in their rendition went
"buzz buzz buzz buzz, buzz buzz buzz buzz, all through the night
buzz buzz buzz buzz buzz buzz buzz buzz,
buzz buzz buzz buzz buzz buzz buzz buzz,
buzz buzz buzz buzz, buzz buz buzz buzz, all through the night"
It is good that they are continuing developing a decent alternative to MS Office but I would rather they spent time on improving the MS Office compatibility than adding in functions for 1% of users.
The redaction tool is something you could achieve the same result with the older versions of Libreoffice with just a couple more steps. Just tried it with Libreoffice 5.1. Export your document as a PDF, open the PDF in Draw, paste shapes over the words you want to redact, select all and convert to bitmap from the tools menu and save it.
"It is good that they are continuing developing a decent alternative to MS Office but I would rather they spent time on improving the MS Office compatibility than adding in functions for 1% of users."
It's meant to be an alternative to MSOffice, not an open source clone. I'd prefer it if MS spent more time on getting decent .odf compatibility rather than working on functions for 1% of user :-)
Nobody is talking about cloning, but if you have to regularly exchange files with MS Office users, it can be a real pain.
LibreOffice's filters for importing or exporting MS Office formats is as bad as MS attempts at importing/exporting ODF files. Either way, neither party shines and the users suffer, if they have to regularly exchange files.
I used to use OO.o and LibreOffice on Linux, but I always had a Windows machine with MS Office in the corner of the office for checking and reformatting documents, before sending them out to customers. In the end, I gave up and went back to MS Office by default, because all my customers used MS Office and I was spending a third of my time checking and reformatting finished documents before sending them out.
it wouldn't surprise me at all if MS switched to ODF at some point – it's a better format – but politics will probably stop that happening. While I do think that, by and large LO has got reasonablly compatible with OOXML, there are still longstanding bugs in some of the code that the developers seem to have little interest in fixing.
They did try, but ODF doesn't provide the flexibility to store current documents created using OOXML - there are a lot of tags and features missing and ODF wasn't interested in adding those features for compatibility, so MS went with OOXML.
That also accounts for why some things go bat shit crazy when imported into LO, because LO is written to handle ODF and those extra tags in OOXML can't be translated into anything useful in LO.
The simplest problems are pagination, which LO generally messes up in Word documents - annoying at best, but embarrassing if you have TOCs and Indexes in your documents that suddenly don't match up any more.
At one employer, they used LO in the conference room and a customer turned up with a PowerPoint slide stack and about halfway through the discussion on process flows, we noticed that LO had moved the flow arrows to random positions on the slides...
Yes, I've heard from people working on standardisation that ODF itself is somewhat rudderless. MS has been reluctant in the past to support work OOXML but have recently really improved in this respect.
I agree with your assessment: lots of stuff works very well but there's a lot that the developers just don't care about. I work almost exclusively with XLSX so everytime I see worksheets set to 1024 x 1024 dimensions I know they're from LO. Sigh: it's an optional element so better left out altogether.
Coordinate systems in OOXML (used when positioning graphical objects) are a complete nightmare!
I must admit I haven't experimented with the digital signatures option, but the already integrated TSCP compliant facility to classify documents is of interest for us, although it requires more study before we decide if we implement it (we will, for instance, need to test if we can set up pre-classified templates). Apparently it's implemented to the point of refusing to copy classified information to a document classed as less secure.
Now, keep in mind that this is *free* software. Amazing. If it wasn't for our Excel munchers I don't think we'd use MS Office at all.
Why would they make it a bitmap where you can't search on text? What a redaction tool should do is simply to replace the characters being redacted with black blobs (or maybe some lightly shaded grey blobs as the "ink saver" option if it will be printed) but otherwise produce a normal PDF. You only need one freaking character in your "redaction font" after all. OK, maybe three - you should randomly vary between three widths so people can't count spacing to determine the length of a redacted word.
It's even more dumb, if the original document is 448 pages, as it makes the redacted version unwieldy.
Mind you it would probably guarantee that no one (other than a determined journo) would download and read the document; so no one would know what Mueller actually wrote..
Or, since they already have protected sections in place, they could gaved replaced the text with some other font for the protected section....
This way it stays one and the same document for everyone and the person having encrypted the document is the only one to be able to read and modify those redacted & protected sections....
Well, a lost opportunity, or perhaps there were other reasons
"...office workers use thousands of columns..."
I've noticed that a given company might employ maybe a dozen financial staff. Made me curious about the rough amount of data that they're dealing with on a weekly basis. Is it a flood of data? Not exactly...
Between time sheets (rolled up for the week), a bit of procurement, some invoices, etc. Total information to be processed is certainly more than 1kB, but way less than 100kB. So very roughly 10kB of actual data (pure info). That's per week.
Putting it in this perspective, what the heck are they all doing? This amount of data could be chiseled into granite at that tiny trifling data rate.
It's all very strange.
"what the heck are they all doing?"
The same thing as the rest of corporate middle management. Creating Power Point presentations and attending meetings. Really. It is typically very easy to get rid of over half of all any given company's middle management based on how many hours of their working week is spent wasted on those two things. And get this ... after the purge, the company runs cleaner, smoother and faster, with zero drop in production. Try it. You might like it.
It's an old study, but there's no reason to suggest much has changed: any data computed from a spreadsheet is more than likely wrong. I imagine that this is often self-correcting and that the people who abuse spreadsheets go on to lose their fingers to chainsaws in misguided can-opening incidents before any real harm is done, but perhaps Excel should simply be banned just to be on the safe side.
I use LibreOffic,5.1.6.2. on Linux 4.15, most days, as a author, well I think I'm an author. Never have any serious problems even with alternate headers, footers, page brakes, page numbers, chapter indexing, and all the other formatting that is required by the printers. And believe me, E. books are even worse. Although I must admit one thing, its spell checker in English English, does have some strange ideas sometimes.
Never have any serious problems even with alternate headers, footers, page brakes, page numbers, chapter indexing, and all the other formatting that is required by the printers.
Tell me the name of your printer please. I guess it's one of those PoD ones. No publisher middle man with copy edit services etc.
Just move away from the crappy Silurian epoch MS Word mimic and give us a usable list styling dialog. Oh, and fix those format painter, clear formatting, frame repositioning and gazillion other bijou bugettes from almost as far back while you are at it, thanks.
Then I would not use all those words that need redacting.
You don't have to be able to code to contribute. Documentation always needs work. If you speak more than one language, your translation skills will probably be highly valued. You can Beta test code, and learn to properly report bugs. Volunteer to answer questions in the user help forum. Cash donations are usually welcome ... even something as simple as turning off your adblocker in the specific forum is a help.
But.. I don't think I would ever have a usage for it.
It's always good to see LibreOffice progressing though, regardless of if it is something I use or not.
For the most part, LibreOffice always works perfectly though!
I can honestly say recently with 6.2 there is only one bug I've noticed. That is, if you write a document in Japanese as opposed to English (I know! Who would do such a thing!), after the first fully written page LibreOffice will slow to a crawl if it has to fallback to Japanese characters from a font such as Sans for example. (Other applications are fine on Linux)
Luckily, using fonts made specifically for Japanese such as Ume, resolves this issue and LibreOffice will resume being fast.
So, aside from that easily worked around bug, all good for the many years I've used it!
Any poor students working on their dissertations should give LibreOffice a go. I managed, a long time ago, to figure out the formatting and even created an automatically updating contents. Bit of a faff but the money saved not buying MS Office (especially on Mac) was well worth it. I’ve never used any other office product since as it handles all my requirements. It would be nice if MS Office had more .odf compatibility but we know Microsoft would rather all their formats were the standard (IE6 springs to mind!).
I use it regularly but my biggest LO gripe was trying to do a basic mail merge with an image on the master page (company logo or whatever). About five years ago a bug appeared which broke formatting if there were more that about 20 images in a document - LO generates a mail merge output document so 30 recipients with a single logo on each page would do the job for instance. This was on the bug list and was fixed then immediately broken in the next update ... last time I looked it was still an outstanding bug.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
