OpenOffice.org for Mac OS X Alpha Released!

An anonymous reader writes "Nearly 6 years after announcing a Mac port, OpenOffice.org has released the first release of OpenOffice.org for Mac OS X that can finally run without X11!! An alpha is available for download today, but a lot of help is still needed to make OpenOffice.org available for Mac OS X. The site is very blunt: 'WARNING: THIS SOFTWARE MAY CRASH AND MAY DESTROY YOUR DATA DO NOT USE THIS SOFTWARE FOR REAL WORK IN A PRODUCTION ENVIRONMENT. This is an alpha test version so that developers and users can find out what works and not, and make comments on how to improve it.' Currently missing functionality includes printing, pdf export, copy/pasting, and multiple monitors. That said, if you're interested in participating you can visit the Mac team to figure out how you can help today."

Comments

REALLY READ THAT WARNING MESSAGE

[Chris_Jefferson]

While this is cool, make sure you really read that warning message. This is real alpha. You won't be able to print. You won't be able to cut+paste reliably. As this alpha has been approaching, I had a crash while saving, leaving me with a half-corrupted useless copy of my document.

So have a look, and help submit bug reports, but please don't try using this is your normal editor, or get annoyed it isn't in a full usable state yet, that's why it is called alpha :)

Re:REALLY READ THAT WARNING MESSAGE

[jalefkowit]

You won't be able to print. You won't be able to cut+paste reliably.

Finally! An office suite on OS X that works just like OpenOffice does on Linux! ;-)

Neooffice - differences?

[cyman777]

OK, so I will start the obvious thread:

What are the differences to Neooffice?
Are they working together?

Besides the slow startup I feel Neooffice already has taken that niche, hasn't it?

Re:Neooffice - differences?

[squiggleslash]

This is a genuine native port, NeoOffice uses a Java intermediate layer to present the UI to Mac OS X.

As I understand it, they're not working with the NeoOffice people, there's always been a little friction between the groups.

In time, this project is likely to overtake NeoOffice, simply because changes to OpenOffice.org will always be faster than those in NeoOffice, which is in a continual state of catch-up.

Re:Neooffice - differences?

[jj13]

The main difference is that neooffice apparently runs through java on the mac, which is why it takes forever to load and can be very slow on occasion. neooffice is at best a substitute, until a native version is released (which is what is being announced today).

Let me point out something I noticed on the site. I'm a budding mac developer, and in reading the dev FAQ I saw that openoffice is being ported using the Carbon API. This is the old API that apple introduced for developers to more easily port their old OS 9 apps to the "new and shiny" OS X, back when it was actually new and shiny in 2001. Carbon does NOT allow you to take full advantage of OS X features, and it's use is frowned upon for new projects by apple and the dev community.

I'm guessing most of openoffice is written in C++ (never actually looked at the source yet), and Carbon is based in C++ so I'm also guessing they didn't want to rewrite everything in objective-c in order to use Cocoa (the "new" API that is like....7 years old now?). In any case, I'm disappointed that the team isn't going for a more thorough port that will let openoffice shine against the competition. Maybe I should put my keyboard where my mouth is and learn the intricacies of mac porting, hehehe.

Re:Neooffice - differences?

[squiggleslash]

Not trying to troll here, and maybe you don't even ascribe to the mindset you imply with the above line, but how can you "overtake" a project that's "in a continual state of catch-up"?

When you start off behind in some areas, but not in others, and in the "others" you can only ever be ahead of the rest. OpenOffice.org's Mac port is at an early stage at the moment, so it implements all of the latest version of OpenOffice.org (by definition) but not all of the Mac native subsystem. NeoOffice includes a complete Mac native subsystem, but not all of the features of the latest version of OpenOffice.org. As time goes by, because making OpenOffice.org native is a discrete, completable, task, OpenOffice.org will catch up with the part of NeoOffice it currently lags behind, but NeoOffice cannot catch up with OpenOffice.org (unless OpenOffice.org ceases to update.)

Does that make sense to you?

Re:Neooffice - differences?

[AttilaSz]

You're wrong. "Carbon" is the collective name for all native Mac OS X APIs, see http://developer.apple.com/carbon/. Quartz, Core Data, Code Audio, etc. are all parts of the umbrella technology set called "Carbon". "Cocoa" OTOH is a handy Objective-C object-oriented abstraction layer atop of that, which is supposed to make development of applications easier. In Windows terms, Cocoa is to Carbon as MFC was to Win32 - an OO encapsulation of the API with convenience goodies. But you can program directly for Carbon if you wish, in the end you have the same capabilities available to your code, it just usually takes less time and lines of code to use Cocoa than Carbon directly. Therefore, it is a perfect solution for you app that you build from scratch. If you're however porting an existing app and it's not trivial to sneak in Objective-C into it, you'd probably go the Carbon route. Nothing to frown upon :-) The misunderstanding comes from Apple's advertized "carbonization" of OS 9 apps ("you need to use Carbon to have your apps run on Mac OS X"). What it really meant was - replace QuickDraw calls with Quartz calls in your source code etc. Carbon is *THE* Mac OS X API, not some transitional support layer for OS 9 migration.

Re:Neooffice - differences?

[x_MeRLiN_x]

You misunderstood what the GP was saying. He acknowledges NeoOffice is currently more popular than OpenOffice.org for Mac but NeoOffice is always playing catch-up due to the fact it is a port of OpenOffice.org and they have to wait for changes to the original before converting them over.

Re:Neooffice - differences?

[spud603]

Adobe uses Carbon for its CS3 suite. Microsoft uses Carbon for Office. Game developers (EA) use Carbon
And apple uses Carbon for Finder, a fact that annoys the hell out of me on a daily basis.
Two things off the top of my head that are implemented in Cocoa apps but not Carbon apps: emacs-style text navigation (ctrl-F,ctrl-B, etc) and on-the-fly word definitions (ctrl-cmd-D while cursor is over a word). There are other differences, too, but I only notice them when they don't work in Finder or in Camino (or Photoshop!).
That said, it's a hell of a lot more integrated than Java!

Re:Neooffice - differences?

[larkost]

You have this almost right...

Many of the sub-systems, especially in things like drawing and sound, often have the more robust API written in Carbon, and then some of the Cocoa API's call those APIs while running. But generalizing like you do that Cocoa is built on Carbon is a mistake, there are many sections of Cocoa that have no Carbon at all underneath them.

A better concept of the major MacOS X API's are that at the root of things you have a layer called CoreFoundation that is written in C. This sits next to the APIs taken from FreeBSD (and the latter dangles down into the Kernel space as well). The primitives from Carbon are often found here, but that is not to say that these belong to Carbon. The primitives found in Cocoa are all built around these, and are often interchangeable with them in some regards.

On top of this you have the "Foundation" layer. This one is mostly written in C or a sub-set of C++ (basically the stuff that does not conflict with Obj-C). Many of the "core" services at the heart of the OS are built here, and at the top of this things start to blur with the bottom of the Carbon layer. Services such as Quartz (but not QuickDraw... which sort-of sits on top of Quartz... but that is messy) sit on this layer.

On top of this layer comes Carbon and Cocoa proper. There is quite a bit of messiness with the two of them calling back and forth, and there are some areas (like Quicktime) that have been very slow to get full implementations in "pure" Cocoa. And a lot more that have had real speed penalties for calling from Cocoa.

Carbon's roots go a little deeper (but less so every new version of MacOS X), but Cocoa and Carbon are philosophically on the same level.

Re:Neooffice - differences?

[jj13]

Getting a LITTLE off topic, but thanks to both of the posts clarifying the relationship of Carbon and Cocoa! As I said, I'm the new guy! But a little more quick research finds that a significant enough part of the community has a hard time with the differences as well. A few informative bits here:

http://www.oreillynet.com/pub/a/mac/2001/05/23/coc oa_vs_carbon.html
http://blogs.msdn.com/rick_schaut/archive/2004/02/ 10/70789.aspx
http://daringfireball.net/2006/10/some_assembly_re quired
http://wilshipley.com/blog/2006/10/pimp-my-code-pa rt-12-frozen-in.html

I have much to learn!

Re:Neooffice - differences?

[Creepy]

They originally disagreed over licensing, now they claim it's much more efficient to just write mac code and not have to coordinate with other platforms - basically, they're saying its easier for them to maintain a fork than it is to coordinate with OOo. The bottom line, however, is NeoOffice code is all GPL, so there is no way for them to legally contribute any of their code to OOo without violating the GPL (unless the original writer submitted it). I don't think OOo is going to move to the GPL, so you've got an impasse.

the "6 year wait" is partly because OOo 1.x was incompatible with MacOS X because of the way symbol bindings were handled (I think it was basically a hack, anyway, exploiting a "feature" in most UNIX-based OSes), so the port really couldn't start until 2.0 (which was heavily rewritten). I was involved in another project when 2.0 came out (I believe STLport, which I think I actually got involved with due to OOo, but X.3 had all the STL features I needed, so I moved on), so I really didn't follow the split that started NeoOffice.

Re:Neooffice - differences?

[diamondsw]

Why is there no moderation "-1 Flat Out Wrong" ?

From the horse's mouth.

Carbon is NOT a fundamental API of Mac OS X. It sits side-by-side with Cocoa, and while it DID start out life as a transitional API from classic Mac OS, it is a peer API of Cocoa. In particular, if you can't deal with Objective-C, you'll likely be using Carbon as it's procedural and accessible from C/C++. Both Carbon and Cocoa are built atop the various "Core" API's. Remember that Mac OS X is a very direct descendent of NeXT, and as late as Rhapsody DR2, there was no such thing as Carbon.

Re:Good news

[kosmosik]

> Although the 'normal' version works like a dream on the Mac,
> having it work without X11 is a bit handier.

Well maybe OOo/Mac/X11 itself works well. The problem is that Apple X11 implementation is crap. You actually need to do stuff from like early 90s Linux to make it work with non-US keyboard layout and this is pain. It can be done via some hacking (like editing cryptic text files and so on) but it disqualifies X11 apps on OSX to rest of the world (apart from geeks).

So native version of OOo is always welcomed. Also I would love to see better X11 from Apple.

Neo Office

[Anonymous Coward]

http://www.neooffice.org/

A port of OpenOffice to Mac OS X that uses Java as a compatibility layer.

It _is_ production ready (I use it every day).
Why the OpenOffice people are hostile to this project is something I've stopped
wondering about... today's announcement of the "first" port of OOO to Mac not
using X11 just shows how badly a project hurts itself when it refuses to work
with others

Re:Neo Office

[GauteL]

"Why the OpenOffice people are hostile to this project is something I've stopped
wondering about... today's announcement of the "first" port of OOO to Mac not
using X11 just shows how badly a project hurts itself when it refuses to work
with others "

Licensing. NeoOffice code can not be reused in OpenOffice.org due to their relicensing to GPL from the original LGPL. This is done on purpose from NeoOffice, and the relationship between OpenOffice.org and NeoOffice is that of host and parasite, rather than a symbiotic one.

Re:Neo Office

[swillden]

Licensing. NeoOffice code can not be reused in OpenOffice.org due to their relicensing to GPL from the original LGPL.

This is incorrect. The problem isn't GPL vs LGPL, the problem is that Sun requires the copyright for all significant contributions to OpenOffice.org to be assigned to Sun, so they can sell StarOffice as proprietary code. The NeoOffice developers don't want their code sold as proprietary, and don't want to assign their copyrights to Sun.

Re:Neo Office

[constantnormal]

Try the current 2.1 version.

Much faster, although since NeoOffice is code on top of OpenOffice, it's never going to be faster than OpenOffice.

And the extensive use of Java as a wrapper around OpenOffice was only the original version 1. The current NeoOffice has much more Cocoa than Java. I suspect that's where most of the speed improvements have come from.

The best thing to hope for is that as OpenOffice itself becomes more OSX-friendly, NeoOffice will be able to leverage their experience in providing OSX-to-OpenOffice integration on top of a better-performing OpenOffice -- unless the approach Sun is taking in making OpenOffice more OSX-friendly is to wrap the C++ core in java, in which case NeoOffice should hang back with the OO 2.0.4 release and blow the doors off the "OSX-friendly" official version of OpenOffice.

The NeoOffice guys have already travelled that road, and speaking as a user, I wouldn't want to revisit it.

Re:Neo Office

[808140]

That seems distinctly unfair. Don't the BSD and LGPL people always say that they don't care if people "take their code proprietary" as it were, and that "the code is still there even if someone else improves it and doesn't share back?" Why, just yesterday there were hundreds of comments to that effect on the GPL2 vs GPL3 story!

It's funny though, because it seems that for all their rhetoric about how using BSD and similarly "non-viral" free software licenses is somehow "more free", BSD/LGPL people generally aren't happy at all when people relicense their code. BSD people hate it when their code gets relicensed, ironically especially when that license is the GPL (for some reason, having their code co-opted by Microsoft or Sun bothers them less -- how does that work?) The LGPL is just like BSD, except that it is exclusively GPL-compatible by design. If it bothers you that someone is releasing mods to your LGPL-licensed program under the GPL, why on earth are you even using the LGPL?

Host and parasite -- god, I love it. Talk about double-speak! It reminds me of this great exchange between an interviewer and Theo de Raadt (whom I have the utmost respect for, as it happens, but this attitude is typical of BSD types):

NF: Lots of hardware vendors use OpenSSH. Have you got anything back from them?

TdR: If I add up everything we have ever gotten in exchange for our efforts with OpenSSH, it might amount to $1,000. This all came from individuals. For our work on OpenSSH, companies using OpenSSH have never given us a cent. What about companies that incorporate OpenSSH directly into their products, saving themselves millions of dollars? Companies such as Cisco, Sun, SGI, HP, IBM, Siemens, a raft of medium-sized firewall companies -- we have not received a cent. Or from Linux vendors? Not a cent.

Of course we did not set out to create OpenSSH for the money -- we purposely made it completely free so that the "telnet infrastructure" of the 1980s would die. But it sure is sad that none of these companies return even a fraction of value in kind.

If you want to judge any entity particularly harshly, judge Sun. Yearly they hold interoperability events, for NFS and other protocols, and they include SSH implementation tests as well. Twice we asked them to cover the travel and accommodation costs for a developer to come to their event, and they refused. Considering that their SunSSH is directly based on our code, that is just flat out insulting. Shame on you Sun, shame, shame, shame.

I will say it here -- if an OpenSSH hole is found that applies to SunSSH, Sun will not be informed. Or maybe that has happened already.

That's from this interview with Theo at NewsForge if you want to read the whole thing. But basically, there's this tremendously hypocritical attitude among the most ardent supporters of licenses that are presumably "freer than the GPL". I see nothing wrong with the classic BSD/PD stance: "We don't care what you do with it, no matter what we still have the original copy". I think that's a noble way to look at things. It just seems that in practice, that's almost never how it is. Someone turns around and creates something useful from your code and relicenses it in a way that prevents you from benefiting, and suddenly they're evil, even though that's supposedly a right that you expressly wanted to guarantee to them in the first place!

Re:Neo Office

[FST777]

When someone takes BSD-licensed code like OpenSSH and releases it proprietary, they may do so. No harm done, "we" still have the original. It is a sign of bad morals when they don't do something back though, and that may be said. Which is exactly what Theo did.

Furthermore, when there is a bug in the original code, it is certainly not up to the devs of that code to inform all the deriatives. That would be funny: someone takes your hard work, does nothing in return and people still expect that you take responsibility for that.

BSD is like: here you are, do what you want, as long as you give credit. It would, however, be extremely nice when there is something done in return, for example financial support. This would create a certain symbiose: Sun makes SunSSH, and gives credit and finances to OpenSSH, so that OpenSSH can continue to improve while SunSSH can use those improvements.

What Theo said was this: sure they can use it, but I find it sad that they won't return any favor. He is completely right with that statement.

If OpenSSH was released under the GPL or even the LGPL, either telnet would still be the #1 app for remote control or a myriad of proprietary protocols would exist for that.

Re:Neo Office

[nanosquid]

Don't the BSD and LGPL people always say that they don't care if people "take their code proprietary" as it were, and that "the code is still there even if someone else improves it and doesn't share back?"

I have released under BSD, GPL, and LGPL. When I picked BSD, I did so for practical reasons, like to encourage companies to use the code with less fear. Nevertheless, I strongly prefer for those companies to contribute back. Legally, I have no way to force them, but I'll certainly tell people in no uncertain terms what I think of them if they don't. Likewise, when I pick the BSD, I do so for good reasons, and relicensing the code under the GPL undermines those reasons, which is why I certainly will tell people who do that in no uncertain terms what I think of them.

In different words...

The BSD license is a license that relies on people to behave reasonably voluntarily; if they don't, you have every right to complain, even if you can't do anything legally about it.

The GPL license is a license that tries to force people to behave reasonably through legal obligations; if they don't, you don't just complain, you go to court.

Released?

[reality-bytes]

You know, "released" when applied to software commonly means software which is considered (rightly or wrongly) to be 'production' material.

This however is apparently an 'alpha' which is commonly an early development version, not fit for general consumption and the type of thing you might get from CVS or a daily tarball.

Some developers use the term 'alpha release' as they assume others will know it's just a packaged up development snapshot, then some muppet takes it and runs to press with it.

Warning!

[Rob T Firefly]

'WARNING: This software may crash and may destroy your data Do not use this software for real work in a production environment.

Do not taunt Happy Fun Software.

Re:Warning!

[CockroachMan]

'WARNING: This software may crash and may destroy your data Do not use this software for real work in a production environment. Windows should have one of those in the box..

Re:Used it on MacOSX - switching to google docs

[Anonymous Coward]

not to mention having every single thing you write indexed for public consumption and advertising goodness...

google...you seem to apologizing to you girlfriend.
would you like to

1. order flowers
2. seek counseling
3. look up a local escort service

Thanks to neooffice...

[MMC Monster]

Just wanted to give a thanks to the folks behind neooffice (http://www.neooffice.org/) before all the bashing starts...

Re:A Massive Waste Of Resources

[BruthaMyles]

You are John C. Dvorak and I claim my £5

Not Alpha

[Doc Ruby]

"Alpha" testing is testing by people who participated in the design and/or implementation. Any testing by people not in those teams is, by definition, "Beta" testing.

Alpha/Beta/Release is not a measure of quality or maturity. It just tells who is testing, and their relationship to the software.

Re:Not Alpha

[shawnce]

Actually in many schools of thought those levels have everything to do with quality and completeness.

Alpha = feature incomplete software with bugs, Beta = feature complete software with bugs, Release = feature complete software with ideally very few bugs.

Still needs X11

[AttilaSz]

http://porting.openoffice.org/mac page says:

In order to run the OpenOffice.org you need to have X11 installed.

Okay, so it allegedly doesn't use X11, but you still need to have it installed? I can see how this is a cheap way of getting around crashes because they forgot to remove some X11 dependency, and it's actually acceptable for alpha software, but it's still really, truly far from elegant...

Re:Huh wha?

[xtal]

Latex might provide better rendering results than terminal. Pico away, however! ;)

Re:Good news

[LKM]

The problem is that Apple X11 implementation is crap (...) it disqualifies X11 apps on OSX to rest of the world (apart from geeks).

And this is precisely what Apple wants. X11 on the Mac is for Geeks, not for "regular" users. The existing issues with X11 are intentional.

Re:Used it on MacOSX - switching to google docs

[NickFitz]

Hey! That escort service has the same phone number as the girlfriend!

Re:Good news

[kosmosik]

> And this is precisely what Apple wants. X11 on the Mac is for Geeks, not for "regular" users.

Yeah so maybe just throw out some source code of X11 that barely compiles and you need to fix it yourself. No binary release - then it would be even geekier. :)

> The existing issues with X11 are intentional.

Yeah. :) That is what I love Mac fanatics - if something is broken in OSX it must be intentional. LoL.

Re:Why?

[Dynedain]

Why? What's the point of this? The Microsoft Office Student/Teacher edition can be purchased readily from Apple's online store for $149 for the full version of Office. If you can't afford $149 for your productivity either:

A) Your time isn't worth any money and you should reconsider what you're using it for
B) You can't afford it, so how can you afford the computer that you're using
C) You just have no desire to pay for software and/or hate Microsoft for XYZ reasons.

Or:
D) You aren't a student or teacher, and the few times per year that you need an office suite don't justify a $300 expense.

$300 is half the price of the Mac Mini to run it on. The $600 spent on the Mini goes a lot farther in terms of productivity than $300 spent on MS Office.

Let me explain what I meant

[LKM]

> And this is precisely what Apple wants. X11 on the Mac is for Geeks, not for "regular" users. Yeah so maybe just throw out some source code of X11 that barely compiles and you need to fix it yourself. No binary release - then it would be even geekier. :)

Not sure what you're trying to say here.

> The existing issues with X11 are intentional. Yeah. :) That is what I love Mac fanatics - if something is broken in OSX it must be intentional. LoL.

Labelling people "mac fanatics" because you don't understand their reasoning is pretty cheap. In your defense, I admit that I was unclear in my original post. Let me explain what I meant.

Apple depends on Mac OS X having applications which do not exist on other operating systems. It's a competitive advantage. Remember NeXT? They had a nice cross-platform development library which allowed NeXT apps to run on Windows. Initially, Apple planned to keep this in OS X. It was called "yellow box" ("blue box" was for old Mac apps).

Interestingly, the idea didn't survive. Eventually, Cocoa became Mac only. Why? Because Apple wants Mac-only applications.

Another example is Java. Making Java apps look good on a Mac is hard. Apple wants to discourage Mac developers from using Java to create cross-platform apps. They would rather keep apps Mac only.

And this brings us to X11. X11 is awesome if you want to run all kinds of apps on the Mac, but these apps don't behave like Mac apps. Why? Because if they did, it would be trivial to write Mac apps using X11 and then port them to other operating systems. Apple would rather keep these apps on the Mac, thus they are discouraging the use of X11 for Mac apps.

Do you now understand the reasoning, or are you still LOLing at me?

The 1980's Called, Wants It's Software Back

[gig]

This is a complete fucking waste of time.

A word processor?

You are killing me. A fucking word processor. It is like inviting people to use a back-breaking chair.

Now that we have more than one output medium, it is important to separate content from style. We also have a "universal" text format which is UTF-8 but we do not have a universal style format. If you munge in your styles with your text you are just setting up a situation where a publishing professional is going to have to rip that text back out of there and if you stored it with a funky old encoding good luck on your smart quotes and em dashes.

What would be the point of enabling a computer user in 2007 type type text and apply styles and you don't save their work as HTML+CSS? What is the point? It makes no sense to me.

What is required when you write is to store the actual typing. If you save UTF-8 you can type any character from any language and then later another human can use that UTF-8 text file to instantly "re-type" your work into any publishing system, smart quotes and all. No conversion necessary, no errors introduced. Doesn't matter if they are working in InDesign or Dreamweaver or other, there is simply no defensible argument for not having a single UTF-8 master copy of any kind of writing. You can drop it on a browser to read even 25 years from now, it will be compatible long after you are dead. In the entire history of computing there has not been a word processing format that lasted even 10 years. If you open a Word document from Word 97, that is only 10 years ago, it has to be "converted" (destroyed) when you open it. Good luck with that system here in the 21st century.

If Microsoft tries to sell ice in the Arctic, will open source follow with open source ice for the Arctic?

Movable Type is about 10,000 times more exciting than OpenOffice. I mean, c'mon.

TextWrangler for Mac OS X is free and it has UTF-8, RegEx find/replace that works across any number of files or a whole disk, real-time speller, S/FTP, lots of writing tools, a great find differences, beautiful text rendering, and completely scriptable with AppleScript (macros). Those are the tools that people need to do good writing and create documents that can be used in modern ways, not mail merge and bad fonts.