OpenOffice 641d Released, Next Stop: 1.0 168
Damek writes "In the spirit of the proliferating news about Office alternatives and 1.0 versions this week, OpenOffice.org has released a new version of OpenOffice, 641d, the last planned release before 1.0. They're calling for help in pinning down and eradicating final bugs before they hit the big milestone: "...we would like you to download it, test it, and finally vote on the feature set.""
What about speed (Score:1, Flamebait)
Re:What about speed (Score:2)
because... (Score:1)
Re:Mod the party line (Score:1)
Hehe. Well you see that may be true for you, sir. But if you let a windows box go for a couple years without reinstalling, it takes a good 5 minutes to boot up. In my experience a brand new fresh install of windows and any install of linux boots about the same speed, at least on a Pentium 2 or above. Well that is with the booting of X. If I don't boot to X it's even faster.
Also I see your point about the general feel of windows being more responsive. However If it's that big of a deal for you check out the preemptive/low latency patches for the linux kernel.
Re:Mod the party line (Score:1)
Sure it works great, and it's reliable. But I thought that the origionator of this thread has a good point.
I would like someone to explain to my just why the origional post was flamebait.
Then could someone please explain why, as a person selecting office machine (desktop) operating system packages and and office software suites I should not be concerned by things like program startup time or GUI responsiveness.
Thanks
how about OS-X ? (Score:1)
Anyone know what the status of the OS_X port is ? I know star division was supporting Mac OS.
Re:how about OS-X ? (Score:1)
IIRC, lousy. OpenOffice seems to be quite a bitch when it comes to portability. It doesn't even build on FreeBSD yet.
Re:*BSD IS DYING (Score:4, Funny)
Re:how about OS-X ? (Score:3, Informative)
Re:how about OS-X ? (Score:1, Insightful)
This would be a universal value to all of its customers, That being said i understand now that SUN is charging for the next version StarOffice why a buisness wouldnt jump headlong into helping another buisness out.
but (Score:5, Interesting)
Re:but (Score:5, Interesting)
Miguel de Icaza [ximian.com] too has said that time is better spent on improving OpenOffice rather than working on say Gnumeric (which he wrote part of too).
So, nothing concrete but who knows, maybe Michael wil work on integrating OpenOffice with GNOME some day. Another possibility is that Sun will do the integration after they switch to GNOME (perhaps they could pay Ximian to do this for them?).
Just dreaming out loud here.
Re:but (Score:2)
I could live with that - I use both. But - I apprecitae Gnumeric's lightweight start-up time. OpenOffice is still in the tens of seconds for me, while Gnumeric starts up in a few seconds.
Maybe if they broke out the separate applications...?
Re:but (Score:2)
Re:but (Score:2)
Roger Dodger.
641d startup time is down to about 12 secs on my system (Red Hat on a PIII500 with 256mb). A considerable improvement, but still kind of embarrassing I think. If you could cut that time in half by breaking out the separate apps, it would approach the acceptable range...
Re:but (Score:2, Insightful)
Lots of people seem to think that redundant programming such as this is bad. The truth is that competition is good as long as the products are compatible. So as long a Gnumeric and OpenOffice can open a common file format the fact that we have development time *wasted* on two products doesn't matter both communities compete against each other. This leads to better products because each group tries to do something new to make it better. The problem is when the groups start to hate each other and don't work together when it makes sense to work together.
Gnome and KDE are a great example of this as they are both use the same *basic* idea but have different implementations. Gnome adds something then KDE adds it and the other way vice versa. Most users don't care as long as they interact in the common areas. That is as long as you can copy and paste from X windows to Gnome to KDE to Java most people don't care what you implement it in.
Alcohol and Calculus don't mix. Never drink and derive!
Re:but (Score:2, Interesting)
Did you see how Mozilla got so much better? I've been so busy admiring mozilla's progress that one day it hit me in the face just how wonderful and fast Konqueror is. I did not switch to Konqueror, but I do use it once in a while and I certainly would not mind browsing the web with Konqueror.
The same with KDE. Once upon a time I was stupid enough to consider both KDE and Gnome a total waste (that was back in the GNOME 1.0 ages). Then I've upgraded my computer and fell in love with Gnome. Every time I saw KDE's face I would turn my face in desgust -- I was a GNOME guy! Only recenty have I been able to lift my head and see the Reality: Gnome and KDE are both mature and wonderful projects that have benefitted immensely from one-another. Just like mozilla and konqueror. And I hear that those guys working on gtkhtml are doing some wonderful progress. Am I wrong? There's always room for a third HTML renderer.
That's competition!
Re:but (Score:5, Interesting)
Which is yet another indication that Miguel has lost the plot. Gnumeric is a stunning app that could seriously rival Excel. OpenOffice isn't close to rivalling either Word or Excel any time soon. But Miguel has long ago forgotten the Unix concept of small specialized tools, and is heading towards MS bloat at an alarming pace. OpenOffice is significantly better than it used to be (and light years ahead of StarOffice 5), but it's starting out on the wrong foot, by trying to be an "office suite", rather than a set of apps that work well together with a consistent look and feel. The sad thing is that I remember Miguel from when he was working on the SPARC and MIPS ports of Linux. How the mighty have fallen...
Re:but (Score:1)
Re:but (Score:2)
Hooray for the team! (Score:4, Interesting)
Re:Hooray for the team! (Score:1)
:) just kidding
I really like Open office (Score:2, Interesting)
heated competition (Score:4, Interesting)
if I were the developers working on openoffice, I'd be thankin my lucky stars(no pun inteded) that sun decided to charge for it. with the growing wave of 'open and free is better' I think they can capitalize on it.
As a former BeOS user, I also noticed gobe productive [slashdot.org] made the news. sweet.
Now comes the important part. in a month, I'm switching over to a completely linux system, and I'm gonna need a replacement for Office. so who's it gonna be?:)
Re:heated competition (Score:5, Informative)
OpenOffice looks good, but when I tried it several times during 2001 it was slow and crashed all the flaming time. I'm sure it's improving but I got bored waiting. Therefore:
To replace Word: KWord looks cool, but I couldn't get equations to work properly. LyX [lyx.org] is really nice if you take the time to understand the concepts behind LaTeX and WYMIWYG. LyX especially rocks for editing equations, but it'll do everything else you could want too, and the output is beautiful. Abiword isn't there yet (tables etc.) but might be one day.
To replace Excel: Gnumeric.
To replace Outlook: I actually use IMP [horde.org], a webmail application. I retrieve pop3 email with fetchmail, make it available via IMAP (one of Debian's IMAP packages) and access it with IMP, on apache-ssl for security, from home and anywhere else with an internet connection. Best thing about IMP is it's the fastest email client I've used! I have folders with hundreds, some with thousands, of emails and the likes of Balsa or Evolution can take forever to access them (if they don't crash). IMP takes seconds, and it never crashes! (I use Galeon for my web browsing/ IMP access). The HORDE [horde.org] project of which IMP is a part is actually an entire groupware suite, but I've only used IMP.
PowerPoint: MagicPoint [mew.org] looks pretty good but I've never used it.
Access: Postgresql or mysql should more than meet your needs. There are nice GUI tools available for both.
Best of luck.
Re:heated competition (Score:1)
Gnumeric vs excel? I've got a fast CPU and lots of memory. I've got a 60 month spread sheet with 100 rows. Why aren't calculations instant? Running the same sheet in the dos version of visi-calc and any change is instant. why should a a computer that is 10000x faster than an XT be 10x slower?
Excel (Score:1)
neither is kspread, but they are getting pretty good for simple usage.
I expect that it is only a matter of a few months before it is usable for me
Re:heated competition (Score:1)
still stuck in 2001 my friend. its 2002!
i'd say OO has become quite the zippy beast. 6 seconds for "./soffice" on my K7/1Ghz/256MB... not exactly THE beast... but one of them
Re:heated competition (Score:2)
MySQL and Postgres only implement a small part of Access, and the graphical frontends I've seen are very thin. The closest things I can think of might be some web-based frontends -- which have a lot of benefits, but also feature lousy data and have no WYSIWYG layout editor, among other limitations.
Re:heated competition (Score:2, Informative)
I've read this comments [zork.net] that suggests the GNUe designer is a possible replacement for access.
Re:heated competition (Score:1)
Well... (Score:5, Informative)
I hope that this changes in one of the future versions, but i have the feeling that it won't.
Re:Well... (Score:2, Insightful)
Re:Well... (Score:1)
How can they be close to version 1 ? (Score:3, Insightful)
Linux only? (Score:1)
Re:How can they be close to version 1 ? (Score:1, Informative)
OOo now runs on Solaris Sparc, Linux x86, Linux PPC, Solaris x86, Win32 (XP, NT, 2000, 98).
Ports are nearing completion on NetBSD (x86, ppc).
FreeBSD is in progress and so is MacOSX.
So from a portability standpoint, I'd say this tree is quite portable.
The big issue on BSD is the lack of standard kernel threads, JDK's etc.
Kevin
Re:How can they be close to version 1 ? (Score:2)
The big issue on BSD is the lack of standard kernel threads...
That very well may be the problem. You see, FreeBSD already has POSIX standard pthreads. Even more compliant with the standard than Linux. If OO isn't building because of the threads on FreeBSD, then OO needs to start using the standard. If they can't be bothered with POSIX, then the least they could do is use a good cross platform thread library like Boost or ACE/CCPP.
Re:How can they be close to version 1 ? (Score:4, Informative)
Slow down there hoss.
OpenOffice is quite portable. It's being developed on Linux and Win32 x86, Solaris (both architectures, Linux PPC, NetBSD, and FreeBSD.
Not all of the ports are keeping up with the main tree, it's true. Since it's a volunteer effort you know what to do about that... the tree itself is probably as portable as anything out there.
Re:How can they be close to version 1 ? (Score:1)
Feature set? (Score:5, Insightful)
OpenOffice at work (Score:4, Interesting)
The 641 build is quite stable and complete. Oh - except for that Australian dictionary. Maybe I should go make one...
I'm looking forward to the proposed changes to the toolbars (look under the 'Todo' section on their site). Looks very nice. Maybe it will come with a performance improvement too. Hint, hint!!!
Re:OpenOffice at work (Score:1)
The evils of commercial software. I use Linux at the office and my 400-MHz desktop machine just keeps getting faster and faster.
Re:OpenOffice at work (Score:1)
Re:OpenOffice at work (Score:2)
Open Office is Weak! (Score:2)
"
glibc version: 2.2.4
/tmp/sv001.tmp/setup.bin: error while loading shared libraries: libstdc++-libc6.1-2.so.3: cannot open shared ojbect file: No such file or directory
"
Open Office is a Beta product! (Score:2)
Beta products have been known to have bugs now and again.
The best thing to do when you note a bug is to check and see if it's already been reported. If it hasn't, then you should go ahead and report it.
Complaining does little to make the product better. Reporting (and helping to fix) bugs does much.
Re:Open Office is a Beta product! (Score:2)
Atleast that seems the way it has been in the past.
Mirrors (Score:1)
the release (Linux binaries, Solver tree and sources)?
I'm trying to download it and provide a mirror, but it's impossible yet.
Re:Mirrors (Score:2)
They don't seem to be entirely /.'d. I was able to navigate their site enough to find this URL for mirror sites [openoffice.org], and load the page of mirrors.
Now I'm downloading a copy from Mexico.
Notice that the build642 directory was last touched Mar 9, while the build641d was last touched Mar 29, so it isn't entirely clear which version to get. However, if you are getting the source you should plan to do a cvs update anyway, so it doesn't matter that much.
Re:Mirrors (Score:1)
Re:Mirrors (Score:1)
If linux is what your after then ftp://openoffice.vosberg.be/install641D_linux_int
Re:Mirrors (Score:1)
Re:Mirrors (Score:1)
Re:Mirrors (Score:1)
version number (Score:3, Funny)
6 41b
6 41c
6 41d
Why the version number contained with bra size?
after this is 1.0,
what's next?
1.0PU
1.1
1.1PU
1.2
1.2PU
(PU = Push up)
Where is the slowness coming from? (Score:1)
My theory (call me paranoid) is that there are time loops in there to make the free version worse than the proprietary version
Re:Where is the slowness coming from? (Score:3, Insightful)
the source code for both these products you mention are OPEN SOURCE. if you can show some of these time loops in the free version, we'd love to see it. i'm sure the developers of the software would love to see it as well.
wrt mozilla (Score:1, Offtopic)
Re:Where is the slowness coming from? (Score:4, Insightful)
Simple - it's largely because they're cross platform. This means that they cannot take for granted ANY system services at all. With Mozilla for instance they reimplemented COM (into XPCOM) because only Windows has such a component model. They created XUL (as far as I'm concerned the coolness value makes that worth the effort alone) because at the time there were no robust enough XP GUI toolkits under the right type of license. Qt would have been ideal, but I think there were problems with the legalese.
So they used their kick-ass rendering engine to do the GUIs. But this makes it larger, as all the widget logic has to be contained within the software. I'm amazed Moz is as small as it is.
OpenOffice is the same - they created their own component model, not sure about the widget set, but because they could assume nothing they had to make a lot of stuff pure Windows/Linux/Mac developers can take for granted.
Re:Where is the slowness coming from? (Score:2)
>>How do they manage to make them so slow?!
>Simple - it's largely because they're cross platform.
You hit it right on the money. It's extremely difficult to write complicated programs so that they're efficient on multiple platforms. Differences in the windowing system are only part of it - another big part is how the different OS's deal with multiple threads, file I/O, etc. - what's very fast and efficient on one platform might be quite slow on another.
If anyone's thinking of starting development on a cross-platform program now, you should seriously look at wxWindows [wxwindows.org] - it abstracts the GUI, file I/O, networking, and many other things and runs on Windows, Unix/GTK, and all MacOS's...and unlike Mozilla and Qt, it uses native widgets on all platforms! Unfortunately wxWindows wasn't mature and stable enough a few years ago when Mozilla was getting started, or even longer ago when StarOffice was getting started, so they had to invent the wheel themselves.
Re:Where is the slowness coming from? (Score:1)
That is simply not possible, since the free version is, well, free. Since the entire source code is available, any deliberate slowdowns would be discovered in no time. The source of Mozilla, for example, can be found right here [mozilla.org]. Feel free to go through it if you like.
A more likely reason is that the free version has some debug routines turned on by default which are switched off in the proprietary version.
A wish for screen shots (Score:2)
Re:A wish for screen shots (Score:1)
Re:A wish for screen shots (Score:2, Informative)
sadly, it doesn't matter how well it works (Score:5, Insightful)
Re:sadly, it doesn't matter how well it works (Score:5, Interesting)
The main thing that matters to most people in an MS Office replacement is how well it reads and writes MS Office files. And that's, unfortunately, a moving target.
Agreed on both points. My experience with 641C (win and linux) is that it reads and writes Office97/2000 files with ease. Really large excel files it barfs on, but your normal .doc with graphics, "normal size" xls files, etc. all work great. I was really surpised at how well it writes the files, too.
Maybe...if the M$ is sufficiently.... (Score:2, Interesting)
Re:sadly, it doesn't matter how well it works (Score:2)
i opened up an 80kb xls file (it had one sheet with about 600 rows and 3 columns of voltages values from a picoscope, and then a graph on those values on the same sheet)
That's interesting; I was having trouble with an electronics-related graph too. In my case it was about 8000 rows and 28 columns of data with the related graph. (simulating a cycloconvertor on a three-phase motor)
Maybe it's largish graphs which cause the trouble; I have a similar file without a graph and it comes right up.
Re:sadly, it doesn't matter how well it works (Score:2, Interesting)
For this reason, just this week I convinced 4 co-workers to switch to OpenOffice. "Read and write Office files without supporting Microsoft!" That easy.
Re:sadly, it doesn't matter how well it works (Score:2)
However, if an organization wants to switch to OpenOffice, they only need compatibility with the latest version of MS Office they had been using. From that point forward, the only problem is communicating with others who use MS Office. But if enough organizations begin to make the switch at the same time, Microsoft will get a taste of their own medicine--yes, they'll have to make Office be able to read/write OpenOffice formats. Granted, it's pretty lame when people send around simple text and data encapsulated in complex formats when ASCII would suffice, but I don't see this changing in the near future. At least the trend is towards XML-based formats.
Re:sadly, it doesn't matter how well it works (Score:2)
The major issue here is likely to be handing office files sent to them. But they could still have troble with some of these if they stuck with MS Office.
Re:sadly, it doesn't matter how well it works (Score:2)
Re:sadly, it doesn't matter how well it works (Score:4, Interesting)
So I turned to Open Office 641c. And to my suprise, bullets exported in an acceptable format. Not perfect. I would still like to see improvement in that area. But its close enough for me to continue using OO rather happily.
I'm confused about the version (Score:2)
(Incidentally, neither of the US mirrors are working, but the one from Denmark seemed to work just fine. The links are further down on the page.
Re:I'm confused about the version (Score:3, Informative)
To clear that up OO642 is the first wave for new development and has lots of new code that breaks and things.
If you want what will be OOo 1.0 eventually, simply grab 641d and ignore 642.
Hope this helps,
Kevin
Re:I'm confused about the version (Score:3, Informative)
http://www.openoffice.org/dev_docs/source/642/rel
641D is a step forward (Score:3, Informative)
I grabbed 641D a couple of days ago, and I have to say that I'm impressed! Other than a few fonts that I haven't migrated to Linux, it's done a great job with complex tables and formatting.
The only thing that would stop me from using it as my regular word processor is that I can't figure out how to make it use imperial units (inches) instead of metric.
Re:641D is a step forward (Score:1)
Re:641D is a step forward (Score:1)
Thank God - maybe we can get rid of 1 oz and 2.5 inches soon then.
Metric measurements and software testing (Score:2)
I have long believed that every developer should spend time fielding support calls, just to make 'm feel the pain they inflict on their customers.
It just occurred to me that developers should also be encouraged to switch between localization preferences from time to time. Heck, alternating their printers between A4 and Letter sized paper every week would either take a significant bite out of user frustration, or save acres of trees.
Just a thought.
I was a skeptic (Score:3, Informative)
Just a couple of notes:
1) I find the interface a little (stress "little")clunky, but I'm a long time Office user. But I'd get used to it in about a week.
2) The Document default views are awful. I'm going to see if I can mess with this to make it more livable for me.
3) It opens Office XP Spreadsheets, Documents, and Powerpoints pretty well. I haven't thrown the kitchen sink at it though.
4) 1/2 hour isn't long enough to judge stability. But I haven't had any crashes or oddities yet.
This is a good package so far as I've looked. I'm going to try to work in it for the next few days and see if its good enough to recommend to relatives who need MS Office compatibility.
Hats off to these guys. This is excellent work.
Debian packages? (Score:1)
Re:Debian packages? (Score:2, Informative)
WordPerfect import filters (Score:4, Interesting)
Re:WordPerfect import filters (Score:1)
Re:WordPerfect import filters (Score:1, Interesting)
Unfortunately, the WP file format is so hideously god-awful ugly that it makes Word files look reasonable and straightforward. Like most other formats (RTF, Quark, or XML, for example), the Word format puts the change-this and end-change codes next to or around the words to be changed. You can interpret the document with direct translation of the codes and get pretty close, and errors in one translation have little impact on other translations.
WP format is the equivalent of having a plain-text document with a macro attached that, upon opening, does the formatting recorded the last time. It's all stuck at the end of the file, and a misinterpretation at any step can screw up the entire rest of the document. You can't just write a "filter", you have to write a whole macro language interpreter...
Still Click-to-focus only? (Score:1)
If I use enter-exit focus then all the menus and dropdown selectors disappear as I move the mouse from the menu/selector title to the menu itself.
this seems to happen irrespective of the window manager I use (even happens without a windowmanager).
Has this changed in 641D?
Regards,
Tim
Re:Still Click-to-focus only? (Score:1)
more bloated OSS (Score:1)
Re:more bloated OSS (Score:1)
Yep. Because it's a well-known fact that closed-source software is invariably free of bloat.
First impressions (Score:3, Informative)
I can gladly say that 641D has introduced significant speed increases under Linux. Startup time fell by half; whereas I used to wait 20 seconds to get a workspace, I now wait 10 seconds or less. The interface in general has sped up. Things feel much snappier, far less laggy. Dialogs open faster, new windows open faster, the whole thing feels like the developers spent much of their time between releases on optimizations and speed increases. I'm already very impressed.
The one thing I used to dread about starting up OO was the speed. I don't think I'll have any such worries anymore, as it doesn't seem to bog down the system either anymore - or at least, not as much.
I'm a happy user.
PS printing and paper size (Score:1)
After some searching, I found the answer: edit file
Also note that this is regarding the Linux version of the 641D release (though it probably works elsewhere).
I still haven't figured out how to change the default style rules without creating (and always having to instantiate) a custom template. If anyone knows how, would you please share? Searching google, google groups, OO.o's issuezilla, and OO.o's mailing lists didn't turn up any good results for me.
Re:RPMs (Score:1)
The binaries does not work for me (at least 641c). They quit silently, and I cannot find out why after several times of strace-ing.
Re:RPMs (Score:1)
641d does.
Re:RedHat compatibility (Score:1)
I have the same comment as most, how can they take software and make it so slow? By the time it loads, I either forgot what I was working on, or lost interest, both of which are very detrimental to schoolwork.
I have had quite a few crashes with it, as well. Saving seems to produce errors more often than not, esp when saving as a word document. Also, the tables are a little messed up. If you want to insert a new row, it will have the same alignment as the row above it, no matter what..say for example you want a table with 8 columns. If you merge two cells together in one of those rows and try to make a new row, there is *no* way to split the cells back up again. Very annoying.
As for the feature voting, I vote that they are all removed. I can run ms office XP on my computer without a problem at all, but when I go into linux and run openoffice, it would be quicker to just draw all the letters of my document in gimp.
Just my 2 cents
Re:RedHat compatibility (Score:2, Insightful)
I still fail to understand why people upgrade office machines as often as they do. My IT person at work tells me she has to upgrade the hardware in order to run the latest software and that often requires an OS upgrade.
Why? Because her users have to be "compatible" with other users in the world. She cited several examples of people receiving files from vendors that were in OfficeXP so we had to have OfficeXP to be compatable.
That was the reason that, when she discovered my 486 box with RedHat 6.2 on it while I was on vacation, she replaced it with a brand new 1.2 GHz machine running Microsoft products of the XP line. She claims this to be an upgrade and is shocked at my lack of thankfulness.
She is, and I'm sure you will be, shocked that I am not happy about this. But damnit my computer was doing just what I wanted it to do, and quite reliably. It now takes longer to write a simple memo, to create a simple spreadsheet, even to check my email or find information on the WWW. But I've got a "compatable" configuration.
Not everyone needs or wants the latest and greatest. Because of this I'm glad that they are testing on a version of Linux that has been out there a while, is fairly common, and has proven itself. The results of those tests should indicate that their product will work fine on newer OS. If not, well that's an OS issue.
Re:RedHat compatibility (Score:2)
Re:RedHat compatibility (Score:2)
Sorry but you just sound silly. Maybe that isn't your intention but in all truthfulness that is how you come off to me. I could see your arguement but to me it's idiotic. Maybe that is because my car from the 1980's is dying and I'm getting ready to toss it.
Adios. And if you don't load linux up on that new computer you should go stew some prunes.
Re:RedHat compatibility (Score:2)
If I were you I'd get them to buy one more hard drive (20-30gb are damn cheap these days) and put Linux on that. If you aren't going to get anything back for the Windows license you might as well keep the ability to use it just in case (unless you don't want that ability which I could understand!).
Maybe my day sucked so much that it has changed my perspective a bit... So for what it's worth I can understand where you are coming from and why my previous attitude (which somehow seems to be a bit close to the bushy tailed admin's) is a bit one sided...
Good luck.