Slashdot Mirror
Linux Office Suites
Cowculator writes: "Sun Microsystems will release the beta version of StarOffice 6.0 in October, with the development version already available. This ZDNet article has some more details, including a link to the development version..." Other submitters sent in notes about Gobe Productive and Hancom Office 2.0, not to mention KOffice and the Gnome office applications. As far as I know all of these are lacking the single most important thing, a robust and complete set of import filters for Word, Wordperfect, Excel, Powerpoint, etc.
Just as important, the lack of EXPORT filters! If you're going to send a document to other people, they need to read it too.
I think StarOffice got off to a wonderful start. I'm very concerned about their progress. The next major version will really be a turning point in the industry one way or the other. If it's solid, and it rocks, with great compatability, then there is a great alternative to office. If it's buggy, or doesn't work well with office formats (especially Excel, where it's the weakest), then MS will win. And I'm going off to live on a deserted south pacific island.
Sigh... If I had to bet, it's depressing where I'd probably put my money... Sun's dropped the ball a few times lately.
Tip to the folks working on it: cool object oriented design is neat, but it's usability, stability, and compatability that will make StarOffice a success. Don't try to do things beyond MS Office, just match it on all fronts! Anything else is an esoteric waste of time.
-me-
Love many, trust a few, do harm to none.
It's obviously pretty essential for users to be able to transfer between the office suites in question and MS Office, if the others are to gain any kind of mainstream acceptance. However, most MS Office users don't actually use something like 90% of the functionality. It's the other 10% that's important.
Further, the only really important Microsoft Office applications are Word, Excel and Access. There isn't the same volume of existing data that must be readily accessible for the other applications.
Now, suppose you could get a solid intermediate format covering those basics (something XML-based, perhaps) adopted as some sort of standard by the free software/open source guys, and have all these office suites using it. It then just needs someone to write a single filter for, say, MS Word docs, to convert to and from the intermediate format, and then all the other Office suites can do it.
I can't believe no-one's thought of or attempted this before, but I don't know of any actual examples. Does anyone else? It must be technically possible; at least, if it's not, you haven't got a hope of converting to the format used by any individual free/open source office suite either.
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
My biggest concern (having implemented Star Schedule server for 30 people so far in a 50-employee company) is that no regard at all has been given to the groupware functionality in OpenOffice. I have very few gripes with Star Schedule, but will need to explain why the newest verions of Star Office cannot be used with the Schedule Server.
If someone were to start a project to make a newer better groupware tool for open office (or some other open-source cross-platform tool), I would find a way to contribute (as I think quite a few others would).
Unfortunately it seems as if ogsproject has died.
Maybe if someone took action and said "All groupware discussions will take place on groupware@openoffice.org" or similar, then at least it wouldn't appear on discuss.
Does Sun not care that there are customers of their software who will be left stranded with data in an obsolete server and egg on their face. I hope not.
Don't even try to match it on all fronts, IMHO. As much as MS would have it otherwise, most Office users are only using a very small subset of the functionality available.
If you can support bulletproof import/export of simple Word documents, with basic things like the formatting, cross-references, tables and so on working reliably, you've got 99% of the portability problems solved. The big issue is the number of documents that already exist in Word format, which people will continue to need to read/edit in whatever new format they're stored. Most of those documents don't use super-advanced VBA scripts, half a million text boxes and WordArt.
Now, if you can go one better, and fix the terminally annoying bugs in Word -- cross-references not updating properly and woefully broken bullets and numbering spring to mind -- then you've also got a technically superior product that solves real problems that MS Word doesn't. Add in the silly omissions -- genuine three-part headers and footers, as used by many, many business documents, for example -- and you're clearly winning.
Of course, similar arguments apply to other Office applications, particularly Excel and Access. I'm simply highlighting Word because the issues are likely to be more widely understood.
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
my K6 @ 266mhz with 64mb of RAM.
So you're the cheap non-upgrading bastard that's to blame for the slowdown in the tech industry...
:)
RTF doesn't support tables, embedded objects, headers/footers, TOC, index, etc.
Completely unacceptable for most companies.
Great for simple documents to retain tabs, bold/underline/italic, etc.
Learning HOW to think is more important than learning WHAT to think.
Gobe actually has great import/export filters, but they're even better: They actually developed an API that anyone can write to, so if they port the API and the filters over to linux (which they are apparently doing), then any application can choose to just write to that API and will immediately be able to save or write in any of the M$ formats that Gobe supports.
BTW, this functionality is based on how BeOS does translation for other formats, too, mainly graphics. Linux could really use to take a lesson from this, because it was one of the coolest and best functionalities of BeOS. Hopefully Gobe will port the full API over, not just the filters themselves.
According to this article, the integrated desktop and probably the start button will be gone in version 6.0.
quote
OpenOffice, and its predecessor StarOffice, are integrated office packages and include a word processor, web browser, and spreadsheet tools. In fact, StarOffice 5.2 contained just about everything a desktop user could need, including an integrated desktop. But with the adoption of desktop environments such as GNOME and KDE, future releases of StarOffice and OpenOffice will no longer carry the integrated desktop.
end quote
The above quote is from the following source:
LWN.net
I for one would like to put aside the KDE & GNOME bias that pushes many to adopt this word processor or that.
Our fundamental problem to be solved is a lack of UNIVERSAL and fully functional MS-Office import *and* export filters. At this point, I would say it's the biggest problem Linux users must struggle with (emphasis on "users" here... the administrators must still struggle with Linux's crappy font management, etc).
RTF, HTML, and the "other" semi-formatted languages don't support popular features very well, such as tables and frames. Would YOU export your resume from a Linux app as HTML or RTF, and leave it to Office to render correctly? HR people are the most "clingy to Office types", and if your resume looks shitty - it's YOUR fault not theirs (world is not fair).
If your RTF resume looks bad in Office, *obviously* you are not a good candidate. You show little attention to detail to allow your resume to overlap characters and corrupt text. I've seen Office mangle some RTF docs that look PERFECT elsewhere -- it's an anti-competitive feature of MS Office. RTF documents from Office, re-import perfectly.
SO... to get to my point, we need good filters. The KDE Office and AbiWord folks should get together on the OpenOffice mailing list, and work to make sure the OpenOffice filters are exactly what they need. There's NO EXCUSE for not standardizing our I/O filters now.
As a great example of co-operation between KDE and GNOME applications, look at gPhoto. This started as a Gnome digital camera app, but the code became something better... a standard Linux API for cameras. Now there's a ton of KDE and Gnome apps, all of which run on top of gPhoto.
Just because KDE and GNOME use fundamentally incompatible desktop libraries, does not forgive these folks for not working together on EXTENSIONS to the desktop. We need more success stories like gPhoto, in areas like Printing, Font Management, pretty Wizards for Samba, etc.
I think about the lack of such examples in Linux, and the thought depresses me...
_Scott
When is Slashdot moderation going to favor less frequent "signal" posts, over "dozen posts a day" noise accounts?
As far as I know all of these are lacking the single most important thing, a robust and complete set of import filters for Word, Wordperfect, Excel, Powerpoint, etc.
There's a real good reason we haven't seen this yet. It's the chicken in the egg problem. Before you can have fully capable import filters, you must first impliment the feature set of the app you're inputing from. For example, Microsoft Word has a bunch of features that do not yet appear in most other "word processors". If your word processor doesn't impliment these features, a filter that does is quite useless for your application (except in regards of ignoring things your application doesn't understand).
Unfortunately, before those features are implimented in your own application, you're going to need some more acceptance (to bring more developers on the project). Unless you can say you do what the mainstream needs/wants, you're still an obscur project. *sigh*
Off this topic, one other thing that kind of bothers me is the massive ammount of reinventing the wheel. Now while having many options is good, there are just far too many open source projects that are each trying to create their own robust, fully-featured office suite. Why is the community wasting so much time?
Some of these really should merge and share code more. Or at least, there should be one organization that is dedicated to creating a unified set of the features found in all open source office suite projects. That way, they could create a big set of libraries that do these things... so when the next guy has this reckless desire to make his own office suite... well, you get the idea.
Why bother.