Slashdot Mirror
LibreOffice 4 Released
Titus Andronicus writes "LibreOffice 4.0.0 has been released. Some of the changes are for developers: an improved API, a new graphics stack, migrating German code comments to English, and moving from Apache 2.0 to LGPLv3 & MPLv2. Some user-facing changes are: better interoperability with other software, some functional & UI improvements, and some performance gains."
Until Java can be exorcised, this project is a non-starter.
Evidence please? Java is alive, kicking and screaming. Java 8 is coming down the turnpike. Java isn't going away anytime soon.
I'm starting to think GNU is the problem with "GNU/Linux" these days.
I just pre-paid £140.00 for MS Office on Gnu/Linux! :(
OpenOffice/LibreOffice is like 90+% C++. The Java bits are mostly irrelevant.
For the sake of order on this sadly degenerating News for Nerds site, please add your post to this parent if the essence of your "thinking" is one of the following:
= LibreOffice is not MS Office, therefore it's crap. .NET.
= LibreOffice uses Java, which everyone know is not as fast and portable as
= LibreOffice lacks MS Office proprietary features and misfeatures, therefore it disappoints me terribly.
= LibreOffice doesn't read or write the constantly mutating, rubbish file formats of MS Office the way only MS Office can.
= LibreOffice isn't backed by a large corporation that Only Wants The Best For Me.
= LibreOffice is bloated, and I insist on the lean responsiveness and stability of MS Office!
= LibreOffice doesn't have ribbons to help me not find features that I used to use.
Rich And Stupid is not so bad as Working For Rich And Stupid.
Not seen any evidence of this, other than the lack of security updates for browser-plugin based vulnerabilities on Oracle's part. Out of curiosity, if you mean server side Java as well as client-side java, could you cite references? This is not to doubt there are big issues with the JVM, and the seeming sloth of response of Oracle to some of them - I'm just wondering what the vast amount of server-side infrastructure would do if the JVM were to be EOLed. In terms of relatively high performance, managed platforms, given Microsoft's flight from CLR, what are the alternatives? And why does Oracle seem so indifferent?
Goddamn idiot mods can't handle the truth! The fact is, java must go. It's a toy 'language' to teach little kids about computers. It was never meant to do real computational work. For that you need assembly, or straight up binary.
write once, infect anywhere?
I'm starting to think GNU is the problem with "GNU/Linux" these days.
Did someone up and rewrite it all to Java, because last I heard it's mostly in C++. Java's just used for optional things.
Well, you had better tell that to Google since it is the core language for all Android apps. You seem to be confusing a few vulnerabilities in Oracle's Java Runtime Environment with the entire Java software ecosystem. In general, Linux systems running Libre Office tend to not even use Oracles JRE. Java isn't going anywhere.
Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
I have been using the last pre-fork version of OO. It works fine and mostly does what I want.
Is there any good reason to switch to the latest Libre?
Admitting that you don't understand the difference between a plugin and Java itself brings you closer to actually knowing what you talk about.
These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
OpenOffice is under Apache Foundation now and it is proper FOSS. This activity only dilutes the efforts to develop a FOSS alternative to MS Office. End it. Don't be childish. Thanks.
I think the vulnerabilities are mostly confined to the browser plug in. Not the entire Java runtime.
I'll see your senator, and I'll raise you two judges.
And you do not need the JRE to run LibreOffice.
...or straight up binary.
Unless you're keying the bits with a set of telegraph keys attached to your CPU's data bus, it's cheating!
Ezekiel 23:20
I think that you are correct. I know that the way I said it nobody can say I misinformed anyone. I stated it that way I did to avoid the case where I might be more explicit and then have misinformed someone.
Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
Java is dead?
Last time I checked, enterprise shops are still hiring more Java developers than any other kind. There are a lot of reasons I don't care for Java, but I would never in a million years say Java is dead.
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
Great, now I will know what the function with the following comment does:
"Gott vergib mir, das ist eine schreckliche Hack!"
And, as we already know, this should speed up builds because your US-made compiler won't have to consult the German dictionary all the time to find out what all the texts actually mean.
Ezekiel 23:20
OpenOffice/LibreOffice is like 90+% C++. The Java bits are mostly irrelevant.
To be precise, as computed by sloccount, libreoffice-4.0.0.3.tar.xz contains:
cpp: 3990644 (87.04%)
java: 400958 (8.75%)
ansic: 91036 (1.99%)
perl: 42456 (0.93%)
python: 17392 (0.38%)
sh: 17256 (0.38%)
yacc: 8228 (0.18%)
cs: 6648 (0.15%)
asm: 3269 (0.07%)
objc: 2602 (0.06%)
lex: 2030 (0.04%)
awk: 907 (0.02%)
pascal: 800 (0.02%)
csh: 235 (0.01%)
lisp: 115 (0.00%)
php: 104 (0.00%)
sed: 7 (0.00%)
However, as Desler said, the Java bits are actually optional.
Note to ACs: I usually delete AC replies without reading them. If you want to talk to me, log in.
posting AC since I don't have a /. account.
in fact, not "code-comments" in LibreOffice were german. Rather it were "germanisms" in the code itself. I can't remember specifics ATM, but it was naming of variables and functions that "looked akward" to native english speakers.
I'm not aiming for informative modpoints, although appreciated :)
I know the specifics because I speak english & german and considered attacking this bug 1-2 semesters ago.
And for a shot at starting a "my license is freer than your license flame war":
BSD/MIT
Fixed for completion's sake. :p
Do what thou wilt shall be the whole of the Law
Why didn't they write a little Fortran, Cobol and D to round the number of languages to an even twenty?
Ray Seyfarth, ray.seyfarth@gmail.com, http://rayseyfarth.blogspot.com
Grammar checking is always fuzzy in a computer algorithm. And because people do not understand grammar (especially in pure English) they rely on word processor nonsense.
That is why it is always crap, and perhaps also explains why (especially young) people cannot do correct grammar (i,e, correct grammar is a bit hard to learn, like it takes work - but letting the computer do it [wrongly] is easy).
To be fair, I dropped this into MS Office, and followed the suggestions until Office stopped flagging:
."
"This here grammar check don't work none
"This here grammar check don't work any."
"This here grammar checks don’t work any." (MS Office says this is correct)
Clearly, LibreOffice sucks donkey balls.
Because it's a spelling checker not a grammar checker.
You dont etch magnetized disks, unless you wish them to cease functioning.
if you read the article and click through to the Sophos website that describes the Mal/JavaJar trojan, you will realize that this is another instance of the browser plugin being exploited and not Java itself.
These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
As someone who did a lot of those translations I may be able to explain the necessity for these:
First, LO's source code is old and massive. Most of the comments I translated were written somewhere in between 1999 and 2004.
Second, as anyone who has ever dealt with old, mature, complex code bases will tell you: you need as much information as possible about your code. This is due to bug fixes and quirks that evolved that code over time (=maturing a code base).
Of course, many of the comments are simple and obvious, but there's always the suspicion that some of them contain crucial information. This is exactly the reason why we're not using software translation programmes (e.g. Google Translate), but rather translate all comments by ourselves.
Finally, please note that the remark about translated comments is not meant as a feature advertisement for end-users, but as a public sign of gratification for those many volunteers that put much of their free time into doing them.
They probably had a surplus of parentheses that they needed to stash somewhere.
You had two rocks?
I can't speak for everyone else, but I was answering questions and testing bugs until 11am. Then I was very tired and got a couple of hours of sleep.
The Infrastructure team was trying to migrate several of the websites over to a new server about the same time as 4.0 was released. After some brief downtime, everything pulled through. Due to a perfect storm of problems during the previous two days, the server upgrades were delayed all the way until release day (oops!)
If you grabbed 3.6.5 this morning, you didn't miss a new release on that branch -- 3.6.5 came out on Jan 30th, and 3.6.6 won't come out until the 2nd week of April. The 4.0.0.3 release is working pretty well, running into a few bugs and issues, but we're working to iron those out as quickly as possible.
Feel free to Ask questions or Report a bug if and when the fancy strikes you.
coding is life
Then again, English is a pretty strange Indo-European language. It has a lot of complexity where it doesn't really add anything, like the plethora of irregular verbs, or the many words that end with the letter e for historical reasons, despite it not being pronounced for centuries. And in other areas, the simpler rules of English come at the cost of expressive ability. Almost non-existent verb conjugation makes things simple, but it also requires 3 words to say "we will run" as opposed to a heavily conjugated verb like "correremos".
Compulsive linguistic fetishist chiming in here:
The "irregular verbs" are not irregular in English (at least, not if you're referring to the stem changers; there are some true irregulars but not many). Swim-swam-swum, sing-sang-sung, etc. (and several other varieties of stem-changers) are ALL regular. They were mislabeled as irregular by 19th century prescriptivist grammarians who didn't know what they're talking about (and who thought that Latin was "perfect" and that any deviation from Latin represented grammatical corruption). The "irregular" verbs are Anglo-Saxon strong verbs. They follow clear, regular patterns and pre-date the influence of Norman French.
Spelling peculiarities are a product of the (relative) freezing of orthography with the invention of the printing press. This is not a linguistic issue, it's one of editorial culture. We COULD have changed spellings as pronunciations changed (and other Indo-European cultures did change their orthography as pronunciation changed in the centuries since the invention of the printing press), but for political reasons have not. It has nothing to do with the language itself (other than the fact that freezing orthography tends to retard language change).
Verb conjugation was not nonexistent in Anglo-saxon words (your "irregular verbs"), and its slow disappearance is the result of Norman influence. Conjugation in Anglo-Saxon was a stem-change, not an inflection. Initially, the only nonconjugating words were borrowed from Latin and Norman French. Over time, because of the prestige status of Norman French, those words became the new normal in English, and old strong verbs begun to lose their conjugations. As an example, it used to be Climb-clamb-clumb, not Climb-climbed-climbed.
Your "we will run" vs "correremos" point about the Spanish being shorter is silly. The English version is 3 syllables and the Spanish is 4. So which is actually shorter to say?
English is weird for an Indo-European language because it is actually a hybrid of the Germanic and Romance branches. This hybridization stripped out incompatible features between the two source families.