OSNews

How about rather than attempting to create an open-source clone (or as good as) Microsoft Office, we actually stop and think about what people really want from a piece of office software?

Personally I avoid using Microsoft Office wherever possible since I find it to be bloated and unpredictable, and often irritatingly designed. Unfortunately, I am yet to find anything better for spreadsheets than Excel, so I have to stick with it for that.

As a good example, I feel people should take more notice of things like iWork. Take, for example, Keynote. Apple basically looked at what people actually want from a presentation program and tried to create it. They didn't just decide to build a replacement for Powerpoint. As a result, I feel that Keynote is much more natural to work with than Powerpoint.

Indeed, one business in my home town runs a large internal Windows network. However, they felt that Keynote offered such a productivity boost over Powerpoint that they actually purchased a few Macs just for running Keynote on.

Now, I am not saying that we should be rushing to recreate the way Keynote and Pages do things (although I am dying for a spreadsheet app that uses iWork's "Inspector" approach). Instead I am saying we should be looking at better ways of improving the usability of such applications from the start. Don't try and re-create Word and then improve it. Start afresh.

For those people who need the really advanced functions and features that Microsoft Word is capable of providing, Word is probably what they want, and you'll have a hard time changing that.

However, there are plenty of people who only use a fraction of the capabilities of Word and would be much better suited to a simpler approach that a fresh, new program could provide.

Fix bugs and code new features. Write documentation and ensure that the Help and API is acessible, discoverable and complete.

Provide Wiki, IRC and web support for users who need help.

Provide beautiful artwork and UI components so OOo looks great to the uninitiated.

If you can't do any of that, and just having a pseudo-discussion/flamewar on a blog is all youre prepared to contribute, then i'd say just wait until someone else does it and use MS Word in the meantime.

I think people see office software as boring. Mostly because OO.org seems bent on reimplementing Office. It's not sexy because it's not innovative and so it doesn't attract developers.
Of course, innovation wouldn't attract Joe user to Linux. We've already proved that with environments, and other programs. It's "familiar" that brings them over.

The codebase is huge; perhaps development should be shifted to a more modular or componentized development paradigm? Such an approach might make OO.org's components more accessible to new developers...which could breath new life and perspective into the project.

Aside from that, I don't see how OO.org can become more "innovative" per say. OO.org seems, by design, more of an attempt to give Windows users a familiar environment to facilitate a "jump ship" from Microsoft products... or allow them to save a couple hundred dollars on an Office suite at least. It's doing a decent job on both fronts.

Unrealistic as it may be, I would like to see OO.org be ported to GTK or Qt entirely. My only pet-peeve in the linux/free*nix world are the 30 bazillion toolkit libraries that I need to have installed/running in memory to have a comparable desktop system. Ideally, I'd like to keep it to only 1 heavy-hitter toolkit... but I realize a port of OO.org to any toolkit would take forever and is largely unrealistic. (Yes, i am aware of the Qt/GTK tie-in projects that try to make it look as native as possible...but it doesn't change what's going on *behind the curtains*...)

I think that OO.o needs more exposure. Right now, it is a niche market. Not because of the features (which are decent), but because most people haven't heard of it!

People go to the store and buy MS Office, or click the MS Office upgrade option when ordering from Dell.com. They have used it before, they know it, and everyone sends attachments with files with the .doc and .xls extensions.

I have mentioned OO.o to several colleagues, none of which have everh heard of it. Zero.

With a bigger user base, there will be more usage. More demand, possibly on the likes of Dell to include it as an option (maybe their base option). This will increase the community size and make the product a bit more "mainstream", though quite likely still the "alternative" office suite.

For most all home users, the features are more than adequate. Get the userbase increased, like Firefox did, and remarkable things will happen. Some may even prefer to upgrade to StarOffice (which would make Sun happy)

Why does Sun have to bear the brunt of the development - why don't IBM and Red Hat pitch in? It's all GPL code, right?

For example, OpenOffice.org 2 has a fantastic support for authoring XForms. How isn't that innovative?

Before they worry about Office 12, maybe they should tackle making the font rendering not look like ass? MAYBE make it so that every word is consistantly rendered instead of the 'dancing i' syndrome.

OpenOffice is a huge, monolithic beast with poor performance. It is not good open source software. Nobody wants to hack on it because it's even more daunting to potential contributors than XFree86 ever was. Worse, it won't scale well with the advancement of the rest of the open source desktop. It can't take advantage of the latest UI toolkits and library backends coming down the pipe.

When a developer attempts to solve a problem (poor font rendering for example), he has to reimplement work that is available through other projects. The community overcomes its relative shortage of developers in part through effective use of modularity and code reuse. It's a shame that the productivity suite of choice for open source desktops doesn't do the same.

Remove the java dependency.

Related to this. I've tried to open/ create macro in Python. This requires java! I do not understand. Why is java a necessary tool for using Python? I'm really confused. Other question: is there some kind of "Python binding", module?, to generate OO.o documents?

i think if 'the community' spends 3 manyears on Koffice, it would be much more effective than spending 30 manyears on Openoffice.

If KDE has learned one thing, it is that a good framework is important. Gnome recieves more money and paid developers compared to KDE, and has more developers in general - but still KDE not just keeps up, but is generally ahead of Gnome. how? 'standing on the shoulders of giants'. they don't accept hacks, things have to be GOOD.

Openoffice is a mess. they should KILL it, or rework it (and i think working on Koffice would be the most efficient way of getting a good Office to Linux).

Of course, there is place for improvement. Globally, I'm very happy with the OO.o beta2 version. That's why I'm saying thank you to the OO.o developpers.

ma_d:
"I'd like to see the Word processor transform into something non-WYSIWYG."
This is an interesting idea. Something, like a "LaTeX" generating OpenDocument (OASIS). But, please, do not blame OO.o developpers for not doing this.

Related to this question. Does anybody know if there is some work on a TEX to OpenDocument converter ? I would not be suprised, if this is already existing. I did not succeed to find infos.

Gnumeric and Kspread are too slow on big spreadsheets. Say 2000 rows by 40 columns. Unusable. For this you have to use Excel or OO.

Kexi is promising, but if you need a database package with scripts, which most people will, its a non-starter. The database part of OO 1.x is unusable for the non-programmer. Spend the money and get Filemaker.

KOffice as a word processor is far more oriented to page layout than almost any other WP package. If that's what you need, great.

If writing books, neither OO nor MS Word is a sensible tool. Probably not KOffice either. For this, nothing will equal Lyx in terms of stopping the page layout aspects getting in the way of composition.

Fortunately most people don't write books, so Abiword, OO or MSWord or OO will be almost equally fine. Here the choice is on import ability, and OO seems to lead on that in Open Source.

How to help OO? Start by explaining to people what it is and is not good for. And pray that V2 fixes the database issue.

The reality is, the primary market for an office suite is the business sector. One of the main problems that OpenOffice.org has is comaptibility with the huge numbers of existing files, especially Powerpoint presentations. I think that the importance of using Powerpoint in the business world is very much underestimated by OO.o and other advocates of OSS software for office suite-type tasks.

I would love to switch to Linux for my everyday work. I even set up a Gentoo system and used it for more than two years. Ultimately, it was MS Office compatibility that sent me back to the Windows world.

statements like this:
"_x_ is key to the success of the Linux desktop."?

I have found OOo, KOffice, and AbiWord (my favorite), FLwriter, Siag office, all to work exceedingly well. On top of those free ones there's also: StarOffice (which provides more licenced fonts and other goodies, Hancom office, and thinkfree office.

A product that seems to have died for Linux was Gobe Productive sadly.

What needs work is more collaboration with hardware vendors from the major linux distros and put an end to hardware troubles once and for all. Particularly in the areas of printers, wifi, and getting 100% legal dvd compliance.

There seems to be more office suites for *nix than you can shake a stick at in reality. The trouble is people keep wanting *nix to be windows.

What's an article about this doing on OSnews anyway? It's operating system specific really. Slashdot maybe.

templates - yes: birthday cards etc, wizards, whatever to make the use of it simple and get results fast in a few easy steps.

with an API that doesn't suck.