Slashdot Mirror
How Is GNOME Office Coming?
Clyde has written a nice article over at LinuxOrbit about the state of the GNOME office suite. With all the hubbub surrounding the recent freeing of Sun's StarOffice, this is gonna get more interesting. I'll tell ya the one thing that I miss in AbiWord is anti-aliased text. Staring at that horridly pixelated text is hard on the eyes. Between the Gimp, Gnucash, Eazel, Evolution, AbiWord, Gnucash and the like (no, I'm not forgetting KDE, I just haven't used it recently), the application support under Linux is rapidly making it feasible for a desktop user, but we're just not there yet. And it's the little things that get ya.
I keep seeing people do odd things with their 'office suite' programs.*
I wonder whether there are more business centered ways of doing this stuff.
Take wordprocessing. WP springs from the need to type letters accurately.**
But letters also
But despite all this context and workflow, the wordprocessor only direcly enables word entry and typographical formatting. Now add to this workflow the needs of B2B interaction, and you suddenly see that a company has a heck of a lot of data tied up inside a multitude of proprietary Word files.
I think feature bloat is a symptom of the painful fact that most software doesn't do what is necessary for business. Consider a small business. Consider the data they use. Forget the office formats used to store the data. Is there really any need for 200MB software packages to manage this small data?
Come on guys, you're information experts. What is the key information that a business runs on? How does this information flow around, in and out of an organisation? What are the tools to manipulate the flow, and what are the structures best suited for arranging this data?
Will the business desktop simply become a departmental transaction processing server...??
*Like keep their list of address mailing labels in a spreadsheet, or keep lists of amendments to cad drawings in a separate wp document.
** I don't know it's real history. Perhaps someone can fill us in?
Unfortunately, though gnumeric looks pretty good, it cannot handle large files well. I recently tried to do a simple import of a comma delimited file and it took over 40 minutes just to import the data. Excel pulled it in quickly (~1 minute). I am really pushing for at least a few Linux desktops in my company, but unless I can provide a good word processor, spread sheet, and email client, I won't make much headway. This is one area where the code from StarOffice may help.
Well, speaking just for the antialiased font problem, I think Linux is going to see some really good news over the next few months. I've been hacking on high performance aa font rendering using Freetype2, and I've got some working code in the Nautilus CVS, as part of librsvg. All of the font integration code is going to get released as LGPL (librsvg is GPL) and I hope and expect lots of other projects to pick up on it.
Not only that, but the XFree86 render project has all of a sudden picked up some momentum. I have confidence that the designs floating by on that mailing list will lead to nice, clean hardware accelerated aa text soon. While we're at it, we'll fix the XLFD mess and problems like client and server fonts not always matching.
This is a hell of an exciting time for advanced 2D graphics under Linux. Gambai!
LILO boot: linux init=/usr/bin/emacs
Oh boy, I can feel the flames licking at this post. Even still, there's a couple of things I've got to say here.
First off, I've tried several times to use both AbiWord and StarOffice to write a simple document. I never could get past a single paragraph. The fonts are just so horribly unreadable as to make using the end product painful. I know it's neither one of those app's fault. Something simply has to be done about how X deal with fonts before Linux is viable for the desktop.
Oh, and I do enjoy reading the "All I want is a text editor", "EMACS is all you ever need", and "TeX rules". I'll give those folks this much credit, the fonts are at least readable there. When it comes right down to it, I fully darn well expect to have tool bars, formatting functions, and all the wiz bangs without having to read some 300 page O'Reily* book. Ack!
(* no offence meant to O'Reily. I've got me a library of them books here)
Moving along here, I've been wondering a little something about the graphical environment in Linux ever since I first got to playing with it. Why are all the icons and window borders so big? Granted, this isn't really a usability thing, but it sure makes those apps look and feel kinda hokey. A good example of this is Gnumeric. Here we've got this pretty darn nice little spreadsheet program that looks like it was put together with children's blocks. Mind you, I only mean to point out Gnumeric as an example. Almost every app running on Gnome, and to a lesser extent KDE, seem to make horrible use of the screen space. It just has a feel of being very blocky. Folks I've shown my Linux setup to have made similar comments as well.
My last bit of a rant here has to do with HTML editors. Why don't we have any decent ones for Linux yet. No, EMACS ain't what I'm talking about either. Most notably over on NT in my mind is Dreamweaver, which aside from being an outstanding GUI for HTML it's also one hell of a site manager. As someone who not only codes the back end of web sites, but also has to do layout and design not having a tool like Dreamweaver around is a serious handicap.
I also have yet to run across anything that approaches the functionality of HomeSite for getting in at the text level. Again, just being a cool text editor doesn't even begin to replace all the stuff that HomeSite has built into it geared specifically for web technologies. Heck, nothing I've seen on Linux even comes close to HS's PHP hi-liting which in and of itself isn't perfect.
I could probably get by for a long while with a less than stellar office suite. What I can't live without is a less than stellar HTML editing suite. There appear to be some interesting prospects on the horizon in development. Maybe some day someone will get enough of this right to actually get me to go closer to full time to Linux. I am watching for it!
The line must be drawn here. This far. No further.
This is not a problem of kernels, or CPU cycles, or the lack of effective 3D displays. It is really so much more basic, and yet, more challenging.
Even though I have not used a typewriter in a year, and the filing cabinet in my office lays there unused, those two objects are the basis of my computer's design.
We may talk about the paperless office oday, but it is meaningless. We are still document-driven rather than knowledge-driven.
Businesses are fighting this, but it is an expensive war. My company uses a system called Onyx for Customer Relations Management. It is a very complicated and clunky client/server system that cost over $500,000 to acquire, and probably that much again per year to train people and keep it running. Look at ERP systems like SAP or BAAN, these are equally moribund, and yet companies still throw billions at them annually, because they are still the best (only?) option.
People do not think in documents, they think in ideas, but you cannot capture the pulses of neurons and transmit them directly. A document is merely a way to capture this stuff so that it can be stored, retrieved, and transmitted to others.
I work as a product manager, and do everything from talk with clients and users to writing specs and drawing mockups for our developers. Most of what I do revolves around collections of ideas of how a feature should work- "The sort button should be on the right side, and all items should sort in ascneding order..." I spend probably 1/3 of my time just maintaining concurrency between specs for development, marketing docs, training manuals, and management summaries.
Let's say I decide the sort button should be on the left instead- that could mean that dozens of documents need to be altered, even though only one concept has changed. Now think- what if I could simply create a "Sort Order" object, and instantiate it in multiple places: word documents, development specs, and page mockups. If I change it, I change it in just one place, and it either updates automatically or at least tells me what needs to be updated by hand. This would not only save me time, it would prevent bugs and many misunderstandings.
I swear I am learning to code now just so that I can take a stab at this problem. I support Linux because it provides a sized canvas on which many ideas can be painted. Gnome is doing a lot of yeoman work, and I am sure it will be useful, but I would love to see more truly speculative design being done. This is where Linux (or some other free OS) could really revolutionize things, because you do not necessarily have to consider the short-term business imperatives that MS and Apple do. Even if either one of them could deliver a system like the one I have described, I believe it would be a failure, because people are not generally ready for such a thing. A small subset would be, and they will adopt it, and in a short time businesses will realize how much more productive those people can be.
-cwk.
There's a very interesting read about the possible division of StarOffice's word processing module, StarOffice Writer, up here at linuxtoday.com.au. It poses an opinion about the possible future of StarOffice now that Sun has GPL'ed the StarOffice source code.
People interested in GNOME Office might also find this interesting.
Digital Philosopher. Looking for work.
KDE 2.0 is already in feature freeze, and in 6 weeks they plan the official release of 2.0, including Koffice.
Some time ago I downloaded the 1.91 release, and was quite impressed.
On the negative side it still had several annoying bugs (which are quite expected at this phaze of the development), is not as feature reach as Star Office, and what hurts the most is the lack of filters.
On the positive side, it has most features users would probably need, is nicely integrated (I was quite impressed with the KOshell), and has original design that does not try to be MS-office clone.
Sela
The author seemed to be lauding the features, such as scheduling, of Outlook + Exchange. Sure, those features are nice, but they reduce choice. To synchronize many people across a project using MS Lookout, it is necessary to be running Exchange server with it. Similarly, all users must be using Outlook as their mail client to schedule projects together. So what happens to those people that wish to use a different e-mail client? They get left out of the loop. And what happens to those offices that wish not to use Exchange, either due to its broken imap implementations, to its inability to function with non-outlook clients, or to its reliance on NT servers? These businesses cannot use Outlook's scheduling features. This is yet another example of how bundling unlike components (mail, schedule) together inherently leads to lack of choice.
--
Friends don't let friends misuse the subjunctive.
Does Gnome Office just sound like a collection of unrelated gtk/gnome software they're trying to group together to compete with kde's koffice?
It just seems to me that koffice is attempting to be a consistent group of interacting programs; Similiar look, feel, interaction, etc and the gnome people are playing the catch up game all over again. I just hope this isn't the 1.0 fiasco all over again.
I thought that was done by all the companies who adopted Office over other solutions, companies desparate for thousands of interoperable work stations that Got Work Done at minimal total cost. As far as I can tell (from living through it), Microsoft just listened really well to what those companies who actually use general office applications actually wanted.
I'd include these attributes in "what they actually wanted":
- one release and deployment cycle for the desktop instead of one per application,
- one training and familiarization cycle for the desktop instead of one per application,
- one partner company for the desktop instead of one per application,
- you get the drift.
WordPerfect and Lotus had viable office software right up to the day Microsoft got this right - and even though Microsoft had to invest in building the Windows platform layer to get that integration, it turned out to be a worthwhile investment for them. (Heck, they even passed Windows off as an operating system, even though it manifestly was not at the time...)A Linux-based successor to MS Office won't have to build an integrated platform - few would argue that Windows 3.x had anything on either Gnome or KDE. And there are obvious cost advantages available to whoever can use either platform to solve the real issues of office software.
Could someone focus on these issues of simplifying things for large organizations with a Linux-based office suite? Yep. Certainly all the wounded corporate support types out there have some ideas about what sucks about supporting Office - if someone found those things and fixed them, that would be a start.
Would a product that did this capture the relevant mind-share? Quite possibly, though one would have to withstand the full competitive force of Microsoft. We know that Microsoft will trade off anything up to and including the laws of the United States of America to keep large organizations on their customer list - I'm sure they'd try to compete on "not sucking" if someone made them do it.
Is anyone focusing on this yet? Not that I've heard - and I'd be surprised if any open source development project looked up from scratching its own itch enough to try to scratch the generic large organization's itch. To my mind, that's what will be absolutely required to change the office desktop world.
"Consider yourself a member of a virtual corporation with Mr. Torvalds as your Chief Executive Officer." - Linux Advocac